THESiS

lllllllll\llllllllllllll‘lllllll‘Illlllllll

3 1293 017879

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

DETERRENTS T0 PARTICIPATION IN EDUCATIONAL
ACTIVITIES BY MICHIGAN FARMERS

presented by

Mark Francis Hansen

has been accepted towards fulfillment
of the requirements for

Ph.D degree in Agricultural &

Extension Education

JMM/ @M
//

 

 

Major professor

Date 12-29-98

 

MSUis an Affirmative Anion/Equal Opportunity Institution 0-12771

PLACE IN RETURN Box to remove this ched<out from your record.
To AVOID HNE return on or before date due.
MAY BE RECAUED with earlier due date if requested.

 

DATE DUE I DATE DUE

DATE DUE

 

gté§t®1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

me CAIRO/0.010.959.“

 

DETERRENTS TO PARTICIPATION IN EDUCATIONAL ACTIVITIES BY
MICHIGAN FARMERS

By

Mark Francis Hansen

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Agricultural and Extension Education

1999

ABSTRACT

DETERRENTS TO PARTICIPATION IN EDUCATIONAL ACTIVITIES BY
MICHIGAN FARMERS

By

Mark Francis Hansen

No matter how motivated adults are to learn about a given topic, barriers exist which can
deter them from participating in learning activities. Understanding these deterrents can
help educators avoid or reduce those barriers and thereby increase participation. Also, as
their personnel and operating budgets become smaller, educational organizations may be
able to recoup costs through registration fees or other methods of charging for services, if

the cost of participation does not out-weigh other types of deterrents.

This study was performed to expand the research related to the deterrent construct as
described by Scanlan and Darkenwald (1984). The purposes of this study were to 1)
develop a Deterrents to Participation Scale for Farmers (DPS-F), 2) identify farmer
deterrents to participation in organized, non-formal, educational activities, 3) determine
how the deterrent factors vary in type and importance in relation to sociodemographic
variables, 4) examine the importance of cost as a deterrent in relation to the other
deterrent factors identified, and 5) investigate whether the DPS-F could be used to predict

participation behavior.

A Deterrent to Participation Scale for Farmers (DPS-F) instrument was developed using
the Deterrent to Participation Scale for General Audiences (DPS-G) as a template
(Darkenwald and Valentine, 1985), as well as through farmer focus group findings. A
statewide mail survey of Michigan farmers utilizing a proportional, stratified random

sample resulted in 466 valid questionnaires (42%) being returned by active farmers.

Six (6) deterrent factors were identified through Orthogonal Rotation, explaining 50. 5
percent of the scale variance. The factors, as ranked in importance using a Friedman’s
Test, were Time Constraints, Scheduling Constraints, Lack of Course Relevance, Low
Personal Priority, Cost, and Lack of Confidence. Cost emerged as being of lesser
importance than most other deterrents. One-way analysis of variance showed that lack of
confidence was more of a barrier for farmers age 65 and over than for farmers ages 18-34
and 45-54. Farmers with the highest gross farm sales indicated more time constraints than
farmers in the lowest sales level. Farmers with off-farm jobs had more dificulty with
scheduling constraints than those without ofilfarm jobs. Female farmers indicated cost as
a more important deterrent than males. Dairy producers indicated that time constraints
were more of a deterrent than did cash crop farmers, livestock producers, fi'uit growers
and “0th ’ farmers such as Christmas tree growers, bee-keepers, etc. Logistic
regression demonstrated that the DPS-F instrument was not successful in predicting
participation behavior for non-participants. Evidence exists that the type and/or
importance of deterrents may be different for participants and non-participants, and that
participation can likely be influenced by removing certain barriers depending on

sociodemographic group.

Copyright by
Mark Francis Hansen

1999

DEDICATION

To my wife, Nancy,
for her love and support as I completed this study.
To my daughters, Molly and Erin,
for the sacrifices they made during
their adolescent years so Dad
could work on his “book”.
To my father, Edward R Hansen,
a World War II veteran,
and a great Dad.

And to God, who was with me
from beginning to end.

And in loving memory of my late mother,
for convincing me that I “...could do anything
I wanted, if I just tried.”

Vera Frances (Campbell) Hansen
1920-1 990

ACKNOWLEDGMENTS

As the investigator of this study, I would like to thank my major professor and
friend, Dr. Carroll “Jake” Wamhofi‘, Professor Emeritus, who patiently guided me through
my doctoral program. I would also like to thank the members of my doctoral committee,
Dr. Mary Andrews, Dr. Marylee Davis, and Dr. Kirk Heinze, who provided me with
guidance and support throughout my graduate program.

I would also like to acknowledge the support of my friends in AEE and MSU
Extension, especially former classmates Dr. Dave Krueger, Dr. Randy Showerman, and
Dr. Murati Suvedi, who finished long before me, but were always there to provide timely
advice and encouragement; Dr. Frank Brewer and Dr. Fred Whims, who encouraged me
to begin a doctoral program, and remained available to assist me; and Ms. Mary Pierce
who seemingly handled all of the “red tape”. A very special thanks also goes to my tutor
in statistics, Ms. Judy Pfafi‘, who kept me smiling while performing factorial analysis.

And last, I wish to thank my sister Maxine and her husband Chuck, my brother
Rex and his wife Linda, the Warren Tonne Family, and other family members and friends,
who supported me throughout this study in their own special and respective ways — a
shoulder for me to lean on, the wisdom to help me decide, the humor to keep me sane, and

the “wind beneath my wings”.

TABLE OF CONTENTS

LIST OF TABLES ........................ i ............................. xi
CHAPTER

I. INTRODUCTION

Statement of The Problem ...................................... 1
Purpose of the Study ........................................ 10
Research Objectives ......................................... 13
Research Questions ......................................... 15
Null Hypotheses ............................................ 15
Definition of Terms ......................................... l6
Importance of the Study ...................................... 19
Limitations of the Study ...................................... 22
Organization of the Study ..................................... 23
11. REVIEW OF LITERATURE

Adult Motivation to Learn .................................... 26
Farmers’ Leaming Activities ................................... 32
Deterrent Research ......................................... 36
Demographics and Barriers .................................... 48

Gender ............................................. 48

Age ............................................... 50

Level of Education .................................... Sl

Socioeconomic Status ................................. 51
Summary of Adult Education and Deterrents ....................... 53

III. METHODOLOGY

Introduction .............................................. 56
Nature of The Study ........................................ S7

TABLE OF CONTENTS

(Continued)
III. METHODOLOGY (Continued)

Research Design ............................................ 58
Introduction ......................................... 58
Instrument Development ............................... 59
Research Instrument Summary .......................... 64

Review of the Research Questions .............................. 66

Generalintion to Populations .................................. 68

Internal and External Validity .................................. 69
Internal Validity ....................................... 69
External Validity ..................................... 71

Reducing Sampling Bias ..................................... 74
Sampling Error ......................................... 74
Frame Error ......................................... 75
Selection Error ....................................... 76
Measurement Error ................................... 77
Non-response Error .................................... 77

Reliability ................................................. 78

Data Collection ............................................. 78

Correcting for Non-response Error .............................. 80

Data Analysis ............................................... 81

IV. REPORT AND ANALYSIS OF THE DATA

Introduction ............................................... 84

Demographic Summary of the Respondents ....................... 84
Response Rate ....................................... 84
Response by Gender ................................... 85
Response by Age Category .............................. 89
Response by Farm Size Based on Acreage ................... 89
Response by Farm Size in Square Feet ...................... 90
Response by Level of Education .......................... 91
Response by Primary Enterprise Type ...................... 92
Response by Annual Gross Farm Sales ..................... 93
Response Based on Farming Status ........................ 94
Response Based on Off-Farm Employment .................. 9S

TABLE OF CONTENTS
(Continued)

IV. REPORT AND ANALYSIS OF THE DATA (Continued)

Participation Behavior ........................................ 96
Objective #1: Development of The DPS-F Instrument ................ 97
Research Question #1: Factor Difi'erentiation ...................... 1 14
Research Question #2: Deterrent Factors and
Sociodemographic Variables ................................. 1 15
ANOVA’s for Factors and Sociodemographics .............. 1 18
ResultsofANOVA LackofCourseRelevance .............. .119
Results of ANOVA: Lack of Confidence .................. 120
Results of ANOVA' Time Constraints .................... 123
Results of ANOVA: Low Personal Priority ................. 125
Results of ANOVA' Scheduling Constraints ................ 135
Results of ANOVA: Cost .............................. 134
ResearchQuestion#3:TheImportanceofCostasaDeterrent .......... 136
Research Question #4: Predicting Participation Using
Deterrent Factors ......................................... 138
Correlation of Factors and Participation .................... 138
Logistic Regression for Predicting Participation Behavior ........ 140
Open-Ended Items .......................................... 143

V. CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS

Conclusions .............................................. 144
Introduction ........................................ 144
Deterrent to Participation Scale for Farmers ................ 14S
Significance And Relevance of Deterrent Factors ............. 146
Factors and Sociodernographic Variables .................... 147
Factor 1: Lack of Course Relevance ...................... 148
Factor 2: Lack of Confidence ........................... 149
Factor 3: Time Constraints ............................. 150
Factor 4: Low Personal Priority .......................... 152
Factor 5: Scheduling Constraints ......................... 153
Factor 6: Cost ....................................... 154
Ranking Cost as a Deterrent Factor ......................... 155
Predicting Participation Based on Deterrent Factors ............ 157
Open-Ended Response ................................. 159

TABLE OF CONTENTS
(Continued)

V. CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS (Continued)

Implications of the Study .................................... 159
Recommendations .......................................... 165
APPENDICES
A. University Committee for Research Involving Human Subjects ............... 172
B. Survey Instrument ................................................ 173
C. First Cover Letter ............................................... 181
D. Postcard Reminder ............................................... 182
E. Second Cover Letter .............................................. 183
F. Table 36. Correlations Between Source Factor Scores and
Sociodemographic Variables for Farmer Participants .................... 184
G. Table 37. Correlations Between Source Factor Scores and
Sociodemographic Variables for Farmer Non-Participants ................ 185
H. Open-Ended Responses to the Questionnaire ........................... 186
BIBLIOGRAPHY ..................................................... 196

95”.”?

9‘.“

10.

ll.

12.

13.
14.

15.

l6.
17.
18.
19.
20.
21.
22.
23.
24.

LIST OF TABLES

Principal Componfis as Found in Various DPS Studies ..................... 44
Research Questions, Tests, and Related Items ............................ 67
Respondents by Sociodemographic Group, Number and Percent of Total ........ 86
Number and Percent of Questionnaires Mailed and Received by

Enterprise Type ................................................. 94
Deterrent Items by Rank, Mean Scores and Standard Deviation ............... 98
Variable Loadings, Item Means, and Scale Ranks:

Factor 1- Lack of Course Relevance ............................... 103
Variable Loadings, Item Means, and Scale Ranks:

Factor 2 - Lack of Confidence ..................................... 104
Variable Loadings, Item Means, and Scale Ranks:

Factor 3 - Time Constraints ....................................... 105
Variable Loadings, Item Means, and Scale Ranks:

Factor 4 - Low Personal Priority ................................... 106
Variable Loadings, Item Means, and Scale Ranks:

Factor 5 - Scheduling Constraints .................................. 107
Variable Loadings, Item Means, and Scale Ranks:

Factor 6 - Cost ................................................ 108
Deterrent Factors in Order of Importance by Explained Variance and

Cumulative Variance ............................................ 114
Repeated Measures Test to Compare Factor Scores ...................... 115
Correlations Between Source Factor Scores and

Sociodernographic Variables ...................................... 117
Summary of AN OVA: Lack of Course Relevance and

Annual Gross Sales ............................................. 120
Surrunary of ANOVA' Lack of Confidarce and Age ....................... 121
Summary of ANOVA Lack of Confidence and Level of Education ............ 122
Summary of ANOVA Lack of Confidence and OfilFarm Job ................ 123
Surmnary of ANOVA: Time Constraints and Age ........................ 124
Surmnary of ANOVA' Time Constraints and Gross Amrual Sales ............. 125
Summary of ANOVA' Time Constraints and Farming Status ................. 126
Summary of ANOVA' Time Constraints and Primary Enterprise ............. 127
Summary of ANOVA' Low Personal Priority and Gender ................... 128
Summary of ANOVA' Low Personal Priority and Level of Education .......... 129

25.
26.
27.
28.
29.
30.
31.
32.

33.
34.
35.
36.

37.

LIST OF TABLES

(Continued)

Summary of ANOVA Low Personal Priority and Gross Annual Sales .......... 130
Summary of ANOVA: Low Personal Priority and Farming Status ............. 131
Summary of ANOVA' Scheduling Constraints and Age .................... 132
Summary of ANOVA' Scheduling Constraints and Annual Gross Sales ......... 133
Summary of ANOVA' Scheduling Constraints and Fanning Status ............ 134
Summary of ANOVA' Scheduling Constraints and Off-Farm Job .............. 135
Summary of ANOVA: Cost and Gender .............................. 136
Friedman’s Test to Rank the Deterrent Factors: Participants and

Non-Participants ............................................... 137
Friedman’s Test to Rank the Deterrent Factors: Non-Participants ............. 138
Correlations Between Deterrent Factors and Level of Participation ............ 139

Logistic Regression with Chi-Square Statistic: Test for Prediction of
Participation Behavior for Participants and Non-Participants in

Educational Activities ............................................ 141
Correlations Between Deterrent Factors and Sociodemographic

Variables for Farmer Participants ................................... 184
Correlations Between Deterrent Factors and Sociodemographic

Variables for Farmer Non-Participants ............................... 185

xii

CHAPTER 1

INTRODUCTION

Statement of the Problem
No matter how much an individual wants to leam, deterrents can emerge which cause him
or her to ignore or miss a learning experience. It may be that the educational activity is
offered at an inconvenient time or location. There may be a registration fee that is too
expensive to pay or the cost may out-weigh the perceived benefits which the activity will
provide. Sometimes, the content of the activity is not relevant or the level of education is
too elementary or too advanced. And sometimes, the individual either lacks the
confidence or the enthusiasm to learn. Deterrents have existed since the beginning of time

and will continue to exist, yet individuals still desire to learn.

Humans are constantly faced with the decision to participate in learning experiences.
Some of these learning opportunities are required so that an individual can keep up with
the technological demands of the job. Others find that even forms of recreation, e. g.,
”surfing the Internet" require knowledge that was not necessary ten years prior. In
addition, learning new hobbies though educational classes is an on-going option for most

of us. The technological change and recreational opportunities interwoven in the

2

agricultural industry are just as demanding of its members as they are in other industries.
Therefore, farmers are faced with the challenge or opportunity of learning new skills or
refining them, either through reading and other individual learning methods, or through
formal and non-formal group methods of organized education. At the same time, they are
also challenged with over-coming deterrents that may affect their desire or ability to

participate.

Perpetual changes in the United States (U .S.) agricultural industry, have continued to
require farmers’ attention since the first settlers arrived in this country. According to
Forbes (1934), one of the first examples in the United States of farmers receiving
education in farming practices occurred when a Native American Indian named Squanto
assisted the Pilgrims in growing corn. \Vrthout Squanto's help, these early settlers would
not have survived the harsh New England climate. As the Colonies developed, an agrarian
economy was formed. Food and fiber production was required to sustain society;
therefore, farmers sought education and information on producing crops and livestock.
Universities of that era were attended primarily by the wealthy and did not ofi‘er curricula
in agriculture. These two barriers, i.e., the lack of wealth and lack of relevant curricula,
served as major deterrents for farmers to attend college. Farmers learned new methods by
sharing their experiences with each other, and techniques were passed on from generation

to generation.

By the early 1800's, farmers were able to obtain new information by participating in

3

Farmer's Institutes and Farm Bureaus (Scott, 1970). These organizations allowed farmers
to voluntarily come together on a more organized basis to learn and discuss new farming
techniques. This desire and method for learning eventually led to the establishment of
Land Grant Universities in 1862 and 1890 respectively (Vines and Anderson, 1976) .
Agriculture and mechanical arts were ofi‘ered for the first time in the US. by these
institutions. However, the cost of attending college and the requirement for family labor
on the farm still deterred many young peOple fi'om attending. In addition, many farmers of
that era had not completed high school and, therefore, were not academically prepared to
attend college. Thus, the agrarian culture itself served as a deterrent to formal learning

beyond secondary school — a deterrent still evident in the late 1900's (Galbraith, 1992).

A number of deterrents to learning were overcome in 1914, when the Cooperative
Extension Service (CBS) was established. Land Grant Institutions served as the
administrative support base for CBS while county agents delivered information directly to
the farmer (Scott, 1970). Legislated through the Snrith-Lever Act of 1914, CES was
considered the educational arm of the United States Department of Agriculture (USDA) in
transferring technology from Land Grant Universities to farmers and consumers (Scott,
1970). Extension education provided an irnportant supplement to formal education in
creating educational opportunities for the farmer and in removing institutional or
situational deterrents to participation. Land Grant institutions provided formal and
organized college courses in agriculture and the mechanical arts for those who could

attend. The arrival of CES, on the other hand, gave those farmers who could not attend

4
college, a continuing opportunity to participate in organized, non-formal learning activities

(Vines and Anderson, 1976).

Placing the educational activities and the instructors (county agents) at the county level
removed some of the key deterrents to learning for farmers. Rural sociologists later
referred to this system of information transfer as the "diffusion of innovations” model,
which many countries around the world have adopted or attempted to emulate (Rogers,
1983). The diflirsion model still hinges on the participant’s desire to learn but fortunately

for educators, most adults have a basic yearning to gain new knowledge (Houle, 1961).

Over time, many of the institutional deterrents of limited access to agricultural research
and information have been removed. However, other deterrents seem to have taken their
place. Situational and dispositional deterrents such as having someone to watch the farm
while the owner-operator is away, schedule conflicts with farming practices and ofilfann
jobs, the costs of registration fees, or lack of interest in studying or learning still appear to

prevent farmers from participating.

Since the 1940's, agriculture has witnessed a vast expansion of technologies in the
products, equipment, and methods used. The advent of nitrogen to boost corn yields was
an important transition in the agricultural technology boom (Rogers, 1983). \Vrth so
much new technology evolving, private industry and other public organizations joined

CBS in ofi‘ering farmers the opportunity to learn about agricultural products and practices.

5

This has led to numerous opportunities for farmers to participate in both formal and
non-formal learning activities, through both public and private sources. However,
regardless of the type or source of the educational activity, decisions to participate were,

and still are, mostly voluntary, and deterrents exist.

With farmers and non-farmers alike, society‘s pace has determined the need for education.
Merriam and Cafi'arella (1991) state that in pre-industrial times the knowledge that a
person needed to succeed as an adult could be learned during childhood. But as
technology expanded, adults found that additional education was necessary. Today, a
higher percentage of individuals attend college or participate in supplemental courses or
seminars in order to acquire jobs, or to up-date or improve their skills (Merriam and
Cafi‘arella, 1991). This has also been true within the agricultural industry. Organizations
and companies support the farming industry by providing educational programs.
Sometimes entities such as the government require that farmers or farm employees

participate in training programs, e.g., certification in safe pesticide handling.

Farming has also been described as a “way of life” rather than a job or career, primarily
because the majority of farmers live and work on the farm. The culture of farming
includes hard work and long days, ofien seven days per week. The phrase “sunrise to
sunset” has been used to describe the farmer’s work day and, in general, this has been a
valid descriptor. Little time or energy is left by the end of the day for the farmer to take

part in extracurricular activities whether work related, recreational, or educational.

6

There is a major difi‘erence between the farm population and the general public in that the
mq'ority of farmers are self-employed. In most situations, no employer or supervisor
exists to encourage or require participation, or to pay for the farmer to participate.
Farmers do not always have an employee or family member who can assume management
of the farm while the owner is away. Therefore, it is difficult to leave the farm during the
day. As a result, farmers must voluntarily seek out educational activities, either for
recreational, informational, or job-related purposes, and then evaluate whether

participation is possible or beneficial.

This voluntary decision to participate in educational activities is of major importance to
those agencies and organizations which sponsor learning activities for farmers. Many
agencies, associations, and private firms now compete for the farmer's time in providing
educational programs. According to Vergot (1991), farmers in the 1980's obtained
information from a number of sources that included both private and public sponsorship.
A Utah study recently showed that the majority of Utah fruit and small grain farmers
surveyed named both CBS and industry publications as their preferred, on-going method
of learning about integrated pest management technologies (Alston and Reding, 1998).
The study showed that this preference was followed by participation in workshops and
conferences. Self—directed learning such as computer access to information, and obtaining
information through media sources were less important methods of learning but still
appeared as a form of education that competes for the farmer’s time (Alston and Reding,

1998).

7

Regardless of the source of information or method of delivery, educational firms and
organizations must have a clearer understanding of farmers as their audience, and what
barriers exist to deter their participation in educational activities. This is important as the
agricultural industry continues to advance as an industry and as farmers seek quality of
life. It is also of critical importance because education, whether private or public, cannot
take place without some form of physical efi‘ort and financial investment. Sponsoring
educational organizations must realize a return on that investment, and it can only come if

participation is strong.

Much effort has been placed on continuing education for adults in the United States.
Houle published the first significant study on motivational orientation by adults toward
education (Merriam and Cafi‘arella, 1991). Houle's book, W (1961),
discusses three types of learners, i.e., those who are "goal-oriented”, those who are
”activity-oriented”, or those who are "learning-oriented". These findings were used as a
foundation for other researchers. Among others, Shefiield, Boshier, Morstain and Smart,
and Boshier and Collins expanded Houle's research and added to the literature on why
individuals decide to participate in learning activities as adults (Merriam and Cafi‘arella,
1991). According to Tough (1971), learning can take place for adults through many
vehicles including both organized education and self-directed learning, or a variation of
these. In each case, however, the individual adult must make that decision to participate,
or not participate, in the learning activity. While research has been done regarding adult

motivators or desires to participate in these learning activities, very little scholarly effort

8

has been devoted to identifying barriers to participation (Merriam and Cafl‘arella, 1991;

Scanlan and Darkenwald, 1984).

Authors have more recently begun to seriously examine the complexity of deterrents to
participation in learning activities that exist to influence leamers' decision-making
processes. Earlier work by Johnstone and Rivera (1965) identified a list of ten barriers to
participation and arranged these into two categories: 1) external or situational barriers,
and 2) internal or dispositional barriers. Cross (1981) later separated a list of 25 barriers
into three categories which she referred to as situational barriers, institutional barriers,
and dimositional barriers. New studies performed by Darkenwald and Merriam (1982)
added yet a fourth category which they referred to as the informational barrier, and re-
defined the term “dispositional” as “pyschosocia ,” which they felt provided a broader
scope of attitudes and beliefs that deterred participation. These earlier studies took a
somewhat simplified and intuitive view of the barriers that may deter adults fi'om

participating in educational activities.

When Scanlan and Darkenwald (1984), began to examine barriers to adult participation in
learning activities, a new pathway was opened. They perceived the issue of deterrents to
participation as a much more complex construct than that proposed in early studies.
Scanlan and Darkenwald proposed that the deterrent construct was multidimensional,
whereby a number of deterrents combine to synergistically deter participation. Scanlan

( 1982) developed the original survey instrument referred to as the Deterrent to

9

Participation Scale (DPS) to examine barriers to adult participation in formal education for
allied health professionals. Darkenwald and Valentine (1985) then used the DPS as a
template to develop a new instrument which was more conducive to examining the factors
that deter the general public from participation in adult educational activities. This new
prototype was coined ”DPS-G”. Subsequent to that study, a Deterrents to Participation
Scale for Bankers (DPS-B) was developed at Rutgers by Davis (1988) and mailed to a
random sample of 1200 bank managers to examine their participation behaviors. The
same year, Weischadle (1988) used the DPS format to exarrrine deterrents to participation
in educational programs with licensed real estate professionals in New Jersey. A number
of other researchers have since explored different populations using the DPS, DPS-G, or

other modifications of these instruments.

The investigators who have developed and used various forms of the DPS instrument
repeatedly found that 1) a multidimensional construct does appear to exist, 2) that the
combination of deterrents impacting populations or sociodemographic groups are similar
but appear to vary in importance by population and subgroup, and 3) that the DPS
instrument can sometimes be used to assist in predicting participation behavior. Previous
survey instruments keying on motivational constructs had not been useful in predicting
participation. More recent investigations have suggested that much additional research
needs to take place with regard to the multidimensional perspective of deterrents to
participation in adult educational activities (Scanlan and Darkenwald, 1984; Darkenwald

and Valentine, 1985; Weishadle, 1988). In addition, recommendations fl'om DPS-type

10

studies have suggested that additional DPS instruments be adapted to measure deterrents

for distinctive sub—populations.

Assuming that participation in learning activities would increase an adult's knowledge, and
therefore increase his or her ability to accomplish more, it seems very beneficial to know
as much about these deterrents as we know about motivators. And if organizations that
sponsor learning activities possessed this type of insight, they would likely be much more

successful in developing programs better suited to learner needs.

Purpose of The Study
Currently, data are limited which explain and predict participation in educational activities.
This is even truer for the farm population, since research which has been performed on this
topic to—date has primarily targeted non-farm audiences. Many difi‘erent organizations,
both public and private, ofi‘er educational opportunities for farmers, and these
organizations would benefit by knowing more about farmers educational behavior patterns
and the barriers that serve as deterrents to their participation in educational activities. The
DPS methodology appears to be the first model which has shown some success in helping

organizations explain and predict these patterns of behavior.

The purposes of this particular study were to 1) develop a deterrent to participation scale
instrument for farmers, 2) identify barriers which serve as deterrents for farmers who

might otherwise participate in educational activities, 3) examine how these deterrents may

ll

vary by sociodemographic group, 4) examine the importance of “cost” as a deterrent for
farmers, and 5) determine whether this model will allow investigators to predict

participation behavior for farmers.

The development of a Deterrents to Participation Scale for Farmers (DPS-F) is an
attempt to produce a reliable and valid scale with which educators can identify barriers
that deter farmers from participating in educational programs. The DPS-F instrument was
modified fiom the Deterrents to Participation Scale for General Audiences (DPS-G)
developed by Darkenwald and Valentine (1985), and the Deterrents to Participation Scale
(DPS) developed by Scanlan (1982). The DPS-F instrument was developed and used to
identify principal components (deterrent “factors”) that represent the barriers to

participation in educational activities as experienced by farmers.

The deterrent factors, once identified, were correlated with sociodemographic data to
initially identify relationships. One-way analysis of variance was then performed to
examine whether there were significant and meaningful difl‘erences between the type and
importance of deterrent factors based on the sociodemographic variables. This
information can assist educational program planners in determining how to best reach
specific types or categories of farmers by removing or diminishing those barriers which
might deter their participation. Previous studies have shown that deterrent factors do vary
based on sociodemographic group. For this study, sociodemographic groups were based

on gender, age, level of education, primary enterprise type, amount of gross farm sales,

12

and farming status (part-time farming versus full time), and whether the farmer holds an

off-farm job.

Previous deterrent studies revealed the importance of cost as a deterrent to the certain
audiences or subgroups. The issue of cost as a deterrent to participation to farmers will be
important to all organizations and firms ofi'ering educational programs to farmers where
fees are considered. Private industry has been successful in attracting farmers to fee-based
workshops and programs where the sponsoring firm intends to make a profit. International
Extension systems have seen varying success in charging fees or developing fee-based
arrangements for assistance, which have helped them support personnel and operating
costs (Kirley and O'Toole, 1993', Harter and Hass, 1992; Healy and Associates, 1991;
Kirley, 1994; Cooney, 1987; Carey, 1992; Gould and Cooney, 1987; Westermarck, 1987;

Finley and Price, 1994; and Brewer, 1996).

Charging farmers a fee for attending programs could help generate revenues for CBS or
any other organization, and may be acceptable to farmers in the US. if the program meets
the farmers' educational needs. During regional “Ag Interface” meetings with Michigan
farmers, participants at five of the six input sessions stated (without prompting) that they
would be willing to pay for Extension programs if the activities included information they
needed (Brewer and Hansen, 1989). At Michigan State University, a farm financial
workshop was filled to capacity even though participants paid a registration fee

comparable to that charged by private industry (Hepp, 1992). The acceptance of this

13

delivery method supports the position that the cost of an educational program does not
necessarily deter farmers from attending, if the program is relevant to their needs. Ifthis is
true, then CBS and other program sponsors can focus on removing barriers other than
registration fees and benefit from the monetary gains. On the other hand, if cost is found
to be the most important deterrent to participation that farmers identify, then sponsors of
educational programs might be advised to avoid charging a fee and either look for other
delivery methods or other sources of revenues. Thus, another purpose of the study is to
examine whether cost is the most important deterrent to farmers regarding participation in

educational activities.

Finally, the DPS format has been used with mixed success in predicting participation
status. For example, Scanlan (1982) and Weischadle (1988) had success using the original
DPS instrument to explain and predict participation, i.e., differentiating between those
who participate and those who do not, based on the deterrent factors. Davis (1988) and
Essman (1994) were not able to repeat predictive results using this methodology. While
new knowledge regarding the deterrents to participation for farmers will do much to help
educators avoid those barriers, a predictive instrument would provide an even more

valuable tool.

Research Objectives
Two objectives were developed for this study. The first objective was to develop a DPS

instrument for farmers. This objective required testing for validity and reliability but did

14

not require hypothesis testing. The second objective required statistical testing to measure
differences and relationships. This objective included four research questions which were

matched with corresponding null hypotheses. The two objectives were:

Objective 1:
To construct a valid and reliable, farmer-oriented form of the Deterrents to Participation

Scale (DPS-F).

Objective 2:

To use the instrument to identify and examine deterrents for farmers as follows:

a) To identify relevant and significant principal components, i.e., factors, that deter adult
farmers from participating in organized adult education activities.

b) To determine how deterrents may vary with respect to sociodemographic variables such
as age, level of education, type of enterprise, amount of gross farm sales, farming status
(hill-time versus part-time), and whether or not the farmer also holds an ofllfarm job.

c) To determine whether the cost of participating in a program, e.g., a registration fee, is
the most important factor that deters farmers from participating in educational activities.
d) To determine whether the DPS-F can be used to predict participation behavior for

farmers in the “participant” category and those in the “non-participant”category.

15

Research Questions
Four Research Questions were included in this study. The Research Questions for the

study represent the second objective of the study, as mentioned above. The Questions are:

# 1) Using a farmer-oriented form of the Deterrents to Participation Scale (DPS-F), can
source variables or factors be identified that deter adult farmers from participating in

organized adult education activities?

# 2) Is there a difl‘erence in the type and/or importance of deterrent factors for farmers
based on the sociodemographic variables of gender, age, level of education, type of

enterprise, amount of gross farm sales, farming status and on- or off-farm employment?

# 3) Is “cost” the most statistically important factor that deters farmers from

participating in educational activities?

# 4) Is there a statistically significant relationship between the type and importance of
deterrents identified by participants and non-participants in educational activities that

would allow prediction of participation behavior.

Null Hypotheses
The null hypotheses for this study correspond with the four research questions stated

above:

16

W: There are no statistically significant and distinct deterrent factors to

explain farmer participation in educational activities.

W: There is no significant difference in farmers’ deterrents to
participation in educational activities based on socio-demographic variables such as
gender, age, amount of gross farm sales, level of income, farming status, or those holding

or not holding an ofilfarm job.

W3: Cost is not the most statistically relevant deterrent in farmers’

decisions to participate in an educational activity.

W: There is no statistically significant relationship between the type and
importance of deterrents identified by participants and non-participants in educational

activities that would allow prediction of participation behavior.

Definition of Terms
The following terms are used in this study. These definitions are provided in order to assist

the reader, and to help clarify points made within the study.

mm: For the purposes of this study, any person 18 years of age or older.
Although some investigators have used age 16, 17, and 25 as the qualifying age to be

considered an adult, in Michigan the legal age considered an adult is age 18.

l7

Clientele: For the purposes of this study, clientele are those individuals who are the target
audience of agricultural Extension programs, i.e., primarily farmers, or agribusiness

representatives.

Wampum: Educational out-reach unit, usually headquartered within,
or responsible to, the national govermnent of a country. In the United States, Extension is
a part of the United States Department of Agriculture (USDA) with leadership provided at
the state level through the Land Grant Universities. Other synonymous terms used in this

study are Extension, Extension System, and the acronym "CES".

Deterrents: Refers to those institutional, situational, informational, or psychosocial
barriers to an individual‘s decision to participate in an educational activity. A synonymous

term used in the study is the word "barriers".

W: A premise that barriers to participation in educational programs
consist of a complex, multidimensional framework, i.e., adults are deterred from
participation by a set of barriers acting in concert, and these may be synergistically acting

on the decision to participate.

W: Group of deterrent items with a single label that represents one type of
barrier to participation in educational activities, e.g., “Lack of Confidence”. Also referred

to in the literature as a source variable, factor, factor group, and principal component.

l 8
W: The primary Likert-scale instnrment used in the study

which identifies deterrent items to participation in educational activities by the target
audience. Also related to or referred to as DPS, DPS-B (DPS instrument for bank

managers), and DPS-F (DPS instrument for farmers).

mm: For this study, the term “educational activities” refers to organized,
non-formal activities which may be work-related, recreational, or informational. Examples
of educational activities include workshops, tours, conferences, field days, tours, lessons,
and seminars . Formal for-credit course work is not included as an educational activity,

nor are forms of self-instruction. Also referred to as an “educational program”.

W3 Refers to a field stafl‘ person of the Cooperative Extension Service, who
serves as the usual point of contact with clientele at the county or local level.
Synonymous terms are Extension Adviser, County Agent, Extension Educator, and

Extension Field Staff Member.

W: In the US. Extension System, an Extension specialist is typically a
professor or specialist who is located on campus at the Land Grant University. Extension
specialists have a full or partial assignment in Extension to support technical and training
needs of the Extension Agents. Extension specialists are sometimes referred to in the

literature as ”subject matter specialists".

19

Farmer: One who produces agricultural products; also referred to in the literature as a
“producer” or “grower”. Synonymously but more specifically referred to as a fruit

grower, vegetable grower, beef producer, etc.

W: In this study, fee-for-service applies to any service provided by
Extension or another public agency, where fees are charged. Afliliated terms are

User-Fees, Fee-Based, etc; not to be confused with Privatization.

W: Institutions of higher learning, originally created to teach
agriculture and the mechanical arts; provided through the Morrill Act of 1862 and Morrill
Act Amendment of 1890; funds to develop Land Grant universities were made available

through land grants from the US. federal government to the states.

Importance of The Study
Farmers must stay abreast of new innovations and technology in order to survive and this
translates into a need for continuing education. Less than 100 years ago, the United
States was still considered an agrarian society. F arnilies lived and worked on the farm, ate
what they grew, and sold the rest to pay other expenses. The shape ofagriculture has
changed over time. Margins of profit have become narrow, forcing farmers to be more
. eficient and thus, better trained. Bemsten (1993) states that between 1950 and 1987 the
US. population grew by 70 percent, while during the same period there was a 67 percent

decline in farm numbers and a corresponding 82 percent increase in farm size. The 1992

20

Census of Agriculture shows that only 15 percent of the farmers in Michigan have gross
cash receipts over $100,000, and these growers produce 78 percent of all farm sales
(Ferris, 1994). Bemsten (1993) charges that, since 1939, non-farrn factors, technical
change, and economic conditions have all played a role in shaping agriculture. And while
the number of farmers has declined over the years, the need for farmer education has
increased. New technologies and eficiencies around the globe have helped U. S. farmers
stay competitive in a global market, but only if they stay current in learning and adopting
new and more eficient methods. Even operators of small and middle-sized farms have
had to maintain a significant level of knowledge to farm profitably and to stay within
regulatory limits. It is important, therefore, that learning opportunities are provided for
farmers. Organizations which provide learning experiences to farmers must understand
what, when, and how to provide these learning experiences so that farmers can and will
participate. This includes not only what motivates farmers to attend educational

programs, but also what deterrents are in play to discourage them fi'om participating.

According to the literature, very little is known about the barriers that deter adults fi'om
participating in adult educational programs, whether these are formal or non-formal.
Furthermore, even less is known about the deterrents that impact farmers in the U. S., as
no specific studies have been performed on this subgroup. It is important that more
research take place in determining how to best provide farmers with educational
opportunities. This study attempts to contribute to that understanding by identifying the

barriers that deter farmers fiom participating in educational activities.

21

Key industry leaders informed the Michigan Cooperative Extension Service (CES) during
the early 1990's that CBS must enhance its ability to meet the increasing technical
demands of the industry it supports (Irnig, 1995; Ferris, 1994). However, because of
shrinking Extension budgets and expanding technologies, Land Grant Universities are
experiencing larger challenges in maintaining adequate numbers of well-trained, technical
field stafl‘. Extension in Michigan is examining various revenue-generating alternatives to
help relieve these budget challenges (Leholm, 1995a; Leholrn, 1995b). Therefore, another
important aspect of this study is to assess the issue of cost as a deterrent to participation in
educational programs for farmers. This information can be used by CBS, or other
organizations, to assess the possibility of revenue-generators such as registration fees for

selected educational programs to assist in supporting the organization's funding needs.

While the results of this study provide generalization to Michigan farmers, much depends
on the response from fanners in various commodity groups, age groups, and income
groups, etc. Results of this study will assist in determining whether there is a difl‘erence in
the deterrents these varying types of farmers identify, whether the cost of participating is a
major deterrent for farmers, and whether behavioral predictions can be made based on

those results.

Finally, this study will serve as a springboard for other research that examines deterrents
to participation in educational programs. The study may also assist in laying groundwork

for securing supplemental or alternative funding for Extension in the United States.

22

Limitations of The Study
This study focuses on the models developed by Darkenwald and others, regarding
deterrents to participation in non-formal, organized educational activities. The
investigator recognizes that many forces are at work in deterring farmers from
participating. Theories on market failure (W eirner and Vining, 1992), valuation (Hubbard
and Hawkins, 1969), and contingent valuation (Mitchell and Carson, 1989) do suggest
that concepts or variables within each of these arenas can serve as barriers. However,
economic theories and constructs were not within the scope of this investigation. This
study focuses on adult learning theories, and on those deterrents that farmer focus groups

identified as being real to them.

It is also recognized that this study depends on self-report by the respondents and is based
on individual perceptions and values. The farming population in Michigan varies widely,
not only due to the sociodemographic groups stated in this study, but also due to
geographic and cultural differences, varying levels of access to educational activities, etc.
As in other sociological studies, human behavior is difficult to interpret and predict.
Response to the deterrent items in the questionnaire are based on a five-point Likert scale
which may limit responses by individuals. Response to the item in the questionnaire which
addresses participation behavior is limited by the respondent’s memory of activities

attended over a twelve-month period.

The process of developing the instrument also could have led to limitations in identifying

23

potential deterrents for farmers. The deterrent items which were used in the questionnaire
were retained based on farmer focus group findings and based on those items found in the
DPS-G instrument. Farmers in the two focus groups included those who were known to
participate in educational programs and those who were not known to participate. While
concerted efi'orts were made to solicit a comprehensive set of deterrents for farmers, a
difl‘erent composition of farmers may have resulted in a slightly different set of deterrent
items for the questionnaire. Using the DPS-G as a template helped to buffer those

difl‘erences.

Finally, the study is limited to individuals actively involved in farming in Michigan. Due to
the difi‘erences in climate, commodity types, educational opportunities, etc, this study

cannot be generalized to farmers in other states or countries.

Organization of The Study
This study is broken down into five chapters: Introduction, Literature Review,
Methodology, Results and Findings, and Conclusions, Implications, and

Recommendations.

The Introduction describes the importance of continuing education for farmers, and how
deterrents to their participation in educational activities have persisted over time and with
the changing structure of agriculture. The Introduction provides a brief summary of

agricultural education and the deterrents to participation for farmers fi'om the time the

24

United States was first settled to present day agriculture. Also covered is an explanation
of the purpose of the study, why this study is important to the educational research

discipline, and limitations which exist in the study. Terms are defined to assist the reader
in understanding acronyms and jargon which may be unfamiliar outside of the educational

or agricultural arena.

The Literature Review provides a comprehensive examination of the related research on
which this study is based. Included are well known theories and constructs on educational
theory, background on farming and agricultural education, as well as data from recent

research dealing with the deterrents construct.

The Methodology section assists the reader in understanding the scientific design and
methods used. Included are the research questions, null hypotheses, identification of
dependent and independent variables, explanation of expected errors in both internal and

external validity, treatment method used for non-response error, and the types of statistical

analyses.

The Results and Findings section provides the important data revealed in the study.
Included are the results of the statistical analysis of the data which tests the null

hypotheses, and an explanation of the findings.

The Conclusions, Implications and Recommendations section summarizes the findings,

25

provides an understanding of what the findings of the study represent, and relates the
findings to the results of other similar studies. Finally, the investigator provides
recommendations on how the data might be used by educational organizations, inferences
toward the general population, as well as recommendations for additional research on this

topic or related subjects.

CHAPTER II

LITERATURE REVIEW

Adult Motivation to Learn
Before one can fully understand how deterrents impact farmers' decisions to participate in
educational programs, it is important to have some knowledge of why adults are
motivated to learn. Research in adult education has taken many avenues of approach, and

many studies have been published.

One of the earlier studies on participation in educational activities was performed by Cyril
Houle (1961). Houle stated that most individuals have a desire to learn, and the audience
he studied all had a number of commonalities when considering continuing education.
These individuals had ”goals” that they wished to achieve, and they found the process of
learning an enjoyable experience. They also felt that learning was a worthwhile activity in
its own right. However, these individuals varied in their ”major conception” regarding the
purpose and value of continuing education. Houle classified these individuals into three
groups. He referred to them as 1) goal-oriented, 2) activity-oriented, and 3) learning-
oriented. Using this terminology, Houle stated that goal-oriented learners have a goal in

mind when deciding to participate in a learning activity. “With this type of individual, the

26

27

need for learning may become apparent upon receiving a mailer or other notification of an
educational activity being offered. Being goal-oriented, the individual typically
participates in learning situations that will help accomplish the learner's goals. Often, these

goals are related to the individuals work or career path.

Houle's second type of learners, "activity-oriented learners", partake of educational
activities for some ”meaning" other than the content provided in the learning exercise. For
example, loneliness is one reason Houle cites for the activity-oriented learner to participate
in a learning activity. It is an opportunity for the learner to interact with others or make
new fiiends, while doing something productive. Others may be trying to escape fi'om the
boredom of daily routines, or to escape from distasteful situations at home. And still other
activity-oriented learners, Houle states, simply wish to attain college credits, certificates or
degrees, and care very little about the subject matter of the courses. Those individuals
who participate due to cultural or family expectations also are identified by Houle as

activity-oriented learners.

The third type of learner, referred to as ”learning-oriented“ individuals, participate to seek
new knowledge, i.e., ”the desire to know", according to Houle. These learners wish to
increase knowledge of life in general, and the things around them. The learning-oriented

tend to read avidly, not just for enjoyment, but to add to their general knowledge level.

Houle submitted that most individuals likely demonstrate some of each orientation when

28
deciding to participate in a continuing education activity, however, typically one of the

three characteristics would be more dominant.

Participation in learning activities is a decision which is usually voluntary for adults unless,
for example, it is required by an employer. Unlike children, who are required to attend
school or coerced to participate in other learning situations, adults make decisions
themselves as to whether, or when, they wish to learn. Tough (1971) states that about 70
percent of all learning projects are planned by the learner. Maslow (1970) describes this
voluntary desire to learn as one of the many needs that humans possess. He also states
that there is a hierarchy of needs that must be considered. For example, basic life-
sustaining needs such as hunger, thirst, or safety, must be met before humans begin to
worry about less critical needs, such as a need for learning or self-actualization. Maslow

(1970) writes:

"We should never have the desire to compose music or create mathematical
systems, or to adorn our homes, or to be well dressed if our stomachs were
empty most of the time, or if we were continually dying of thirst, or if we
were continually threatened by an always impending catastrophe, or if
everyone hated us. "

These needs, in order of importance, include hunger and thirst, safety, belongingness, self-
esteenr, and the need for self-actualization (Maslow, 1970). The term self-actualization,
refers to an adult's compassion to become what people feel he or she can become.

Maslow states, ”What a man can be, he must be." Learning activities are a part of self-

29

actualization, and learners decide for themselves what activities to participate in to meet
their expectations. Tough (1971) states that an individual may want to participate in some
type of learning exercise just to make an informed decision on purchasing an appliance, or
he may wish to learn something with more long-term impact such as supervisory skills to
use on the job. But either way, it is still a decision by the learner to participate in the
learning activity. Tough states that ". . .many adults make highly deliberate efl‘orts to

learn. "

According to Tough (1971), the average person participates in eight learning projects a
year. In his studies, Tough described a learning project as a learning activity at least seven
hours in length. His research showed that the typical adult in the US spends about 700 to
800 hours a year in learning projects, and some individuals spend up to 2000 hours in
learning projects. He also found that 98 percent of the audience surveyed in one study,
indicated that they had participated in at least one learning project during the past year.
However, Tough's studies have shown that the amount of time spent in learning projects

varies with type of job, social class, gender, age, and educational level.

Cross (1981) also discussed issues related to who, why, and how, adults learn, and also
refers to external forces, i.e. , the ”environment", that impact individuals' decisions to
participate in learning activities. Cross states that there is a current and growing increase
in lifelong learning which can be seen in the adult population. She attributes this to three

influences, i.e. , 1) changing demographic factors with a larger adult populace than before

30
1975, 2) social change that includes the changing roles for women, e.g., early retirement, a
rising education level of the general populace, etc, and 3) increasing technology that
demands continued learning by workers and consumers to keep up with change. Because

of this, adults are choosing to, or are required to, participate in continued lifelong learning.

While each of the aforementioned influences on adult education are important, the larger
numbers of adults in society today compel us to respond to adult leamers' needs.
According to Merriam and Cafl‘erella (1991), the median age of Americans is rising —
from 30.6 in 1982, and will reach 36.3 by the year 2000. This trend toward an increasing
age of society was caused by the post-World War II baby-boom. This group caused a
population expansion that moved through the elementary grades, to middle schools and
secondary schools, than to post-secondary schools and universities. Now these baby-
boomers have reached adulthood. The need for adult education is expanding with the
baby-boom population, just as it did in each of the previous stages (Cross, 1981).
However, according to Cross, adult participation in educational activities in the 1980's
out—distanced even the expansion of the adult population in the U. S.. This indicates that
even though adults have become more numerous, more than just the number of adults is

impacting the increased demand for adult education.

Merriam and Cafi‘arella (1991), as well as Cross, comment that the demand for adult
participation in educational activities has also changed over the years due to society's

expanding knowledge base. In pre-industrial societies, the rate of change was slow and

31

what was needed in adulthood could be acquired in school as a youth. Today, adults must
constamly acquire new knowledge to keep up with changing technologies (Merriam and
Cafi‘arella, 1991). This has been particularly true with farmers. Over the past 20 years,
research has unveiled much more information about plant and animal genetics and
breeding, pest biology and management, information systems related to agriculture, and a
host of other technological advancements. During this same 20-year period, farmland and
farm numbers declined while farm size grew (Bemsten, 1993). It became necessary for
farmers to become more skillful and eficient in their professions to maintain viability. In
order to do this, farmers, like other members of society, have had to learn and adopt new
methods. They also have had to become aware of new products, and in some cases,
develop new philosophies toward their profession. They have done this through many
difl‘erent learning methods including reading, attending farm meetings and seminars,
participating in farm tours and conferences, utilizing media outlets, and by consulting with
private or public consultants. According to Vergot (1991), farmers today acquire

knowledge through many different sources.

For the purposes of this study, farmers are adults who are at least age 18 or older. It is
logical to expect that the same behaviors that afl‘ect the general adult population regarding
learning styles and educational participation, would affect them. However, while learning
orientation, motivators and needs for self-actualization may exist, deterrents that impact

the farmer may differ fi'om those afi‘ecting the general populace.

32

Farmers’ Learning Activities
F arrners like other adult individuals make deliberate decisions to learn. Several authors
(Tough, 1971; Cross, 1981; Merriam and Cafl‘arella, 1991), state that at least part of an
adult's reason for participating in educational projects is to increase competencies in his or
her job. Unlike many jobs in the marketplace, the farmer must be a botanist, an
entomologist, an animal nutritionist, a mechanic, a commodity broker, and often a labor
supervisor. Farmers are rarely able to specialize in all of the roles they play, however, a
breadth of training is necessary to maintain minimal skills. Coupled with this dilemma is
the ever-expanding technology in the agricultural industry (Ferris, 1992). With changes in
the agricultural industry leading to larger, more specialized farms (Bemsten, 1993),
coupled with increased technology and communication, it is expected that farmers will
have to increase participation in educational activities, whether formal or non-formal, to
compete in a global market (Shapley, 1992). Therefore, farmers must continue to learn
throughout their farming years, in order to maintain an efiicient and profitable farm

business.

Learning may occur through participation in educational meetings, conferences, or tours,
or merely through reading magazines, newspapers, and market reports. Tough (1971)
states that "adult reading” is a widespread method of learning for adults in America.
Tough demonstrated in one study, that adults spend at least one hour a day, or over 300
hours a year reading newspapers, with the majority indicating that they read newspapers

for practical use. Only 12 percent of these individuals mentioned that relaxation was the

33

intent. Magazines also were demonstrated to be an important source of knowledge for
adults, according to that study. Vergot (1991) found in a study of Southwest Michigan
fiuit and vegetable growers, that farmers do learn from various sources. Vergot’s study
reported that farm magazines, agricultural bulletins from universities, and weekly or
monthly newsletters ranked at the top of 18 channels of information that farmers use in
order to gain new agricultural information. However, participation in educational
programs such as field days, trade shows, and twilight tours were still important sources

of information.

Johnstone and Rivera (1965) stated that the major emphasis in adult learning was ”on the
practical rather than the academic; on the applied rather than the theoretical; and on skills
rather than on knowledge or information. " This compares with input from Michigan

farmers who have continuously asked Extension for useable and practical information and

research.

Johnstone and Rivera (1965) found that the most significant group of learning activities
indicated by the adults in their survey, was related to topics that dealt directly with daily
tasks and obligations. Farming represents a five-, six-, and often sevenoday work week,
depending on the farming enterprise and personal values. Many have described farming as
“a way of life”, referring to the fact that most farmers live on the farm, and are therefore
continuously involved in the demands of the farm, 24 hours per day throughout the year.

Thus, Johnstone and Rivera’s reasoning would lead us to the conclusion that farmers

34

would more likely participate in educational opportunities related to farming activities than
other learning activities, e.g., recreational activities. Scott (1970) states that, in fact, the
reason for the establishment of Farmers Institutes, and later the U. S. Extension System,
was to respond to farmers demands for practical and useful information related to their

farming operations.

The reasoning behind farmers decisions not to participate in educational programs appears
to be as complex an issue as it is for other adult groups. In preparation for this study,
assessments of farmer deterrents were collected through focus groups consisting of active
farmers. These individuals indicated that a number of deterrents may be in play which
impact farmers decisions to participate in educational activities. Deterrents may include
cost of registration fees that are too high, unawareness that a program is being offered,
apathy toward education or toward attending educational events, conflicts with personal
or professional schedules, inconsistencies between program content and the learner's
needs, unsatisfactory past experiences in attending learning activities, lack of a responsible
individual to manage the farm while the owner/manager is away, educational activities
ofl‘ered during inconvenient times or seasons, educational activities not held when the
farmer needs the information, etc. However, at the current time, scientific data is lacking
on this topic and especially with regard to adults engaged in farming as a profession.
Therefore, organizations that sponsor educational activities for farmers continue to guess

at the types of barriers that may deter farmer participants from attending their educational

programs.

3 5
Extension agricultural agents will point out that certain farmers repeatedly attend
Extension programs throughout the year, while other farmers seldom, or never,
participate. There does not appear to be a trait or situation that is common to either
group. Those who attend Extension educational programs represent both large and small
farms, full-time and part-time farmers, young and old age groups, and farmers with
various levels of education. The content of the program may or may not be a deterrent (or
motivator), however, an era existed in the 1980's when “gasohol” (alcohol made fi'om
farm crops) continually drew large crowds at Extension programs. Perhaps the novelty of
this topic, or the fact that gasohol might create a new alternative market for their crops,
motivated farmers to attend. Regardless, there are still individuals who are quite faithful in
attending Extension programs, and those who are seldom associated. This appears to
indicate that those who attend are motivated by certain aspects of educational programs,
and there are others who are less motivated. This participation behavior could also be
seen from a deterrent perspective, i.e., that some farmers are deterred fiom attending

while some are not deterred.

Ownership, and thus participation, is achieved by including members of the user group
during the planning process (Levine, 1992). Extension agents have attempted to address
this when planning programs and this has been helpful, i.e., utilizing farmer-participant
advisory groups to identify programmatic desires. Extension agents also attempt to keep
costs low, and hold meetings in convenient locations or at convenient times or seasons.

Yet, there are still many farmers who do not participate. This may indicate that program

36

content, cost, location, and timing are neither significant motivators nor significant
deterrents to participation for this group, or that the correct combination of these

deterrents/motivators vary fi'om group to group or individual to individual.

It is also well known by professionals who work in the agricultural arena that cultural
farming practices and traditions are often handed down from generation to generation,
within a family farm operation or cultural group. Houle found that ”carrying on a tradition
of their family or culture" was a characteristic that caused some individuals to participate
in learning activities (Merriam and Cafi‘arella, 1991). In other words, children may adopt
the parents’ perspective toward continued learning, and then pass that value on to the next
generation. Therefore, family members might be expected and/or encouraged to attend
educational activities if that was a family value. On the other hand, negative values
toward continuing education could also be handed down through the family as suggested
by Darkenwald and Merriam (1992). If so, the lack of participation in educational
programs, could be a tendency that has been accepted within a farm family culture, and
therefore it may exist as more of a cultural non-expectation than a personal distaste for
continued learning. Both Scanlan (1982) and Darkenwald and Valentine (1985) examined

the absence of family member support in their questionnaires.

Deterrent Research
During the 1980's a new scale instrument was developed by Scanlan (1982) to investigate

the construct of deterrents to participation in educational programs. Why adults

37

participate in educational programs had been the subject of much earlier research but until
recently, little study had been performed regarding the barriers which deter adults from
participating (Merriam and Cafl'arella, 1991). The foundation for the deterrent construct
was performed at Rutgers University in New Jersey (Scanlan, 1982; Scanlan and
Darkenwald 1984; Darkenwald and Valentine, 1985). Using principal components
analysis, these researchers identified deterrent factors that varied between the difl‘erent
populations studied. These studies viewed the deterrent construct as a multidimensional
phenomenon where various factors worked simultaneously and sometimes synergistically
to deter adults from participating in educational programs. Prior to that period, the issue
of barriers to participation in educational activities was seen as a more intuitive and

simplified construct (Scanlan and Darkenwald, 1984).

Some of the earlier studies to address deterrents to participation in educational programs
were performed by Johnstone and Rivera (1965). These investigators looked at
educational experiences of the adult American population. Participants in that study were
either over the age of 21, married or the head of a household, and had terminated or
completed regular, full-time school attendance. In the study, these investigators identified
motivators as well as deterrents to adult participation in educational programs. Johnstone
and Rivera termed these deterrents as "barriers" and listed financial restraints (or cost),
busy schedules, and lack of physical energy as the deterrents most fi'equently reported by
respondents. They reported that women identified more obstacles to enrollment than men,

older adults reported more obstacles than younger adults, and persons in lower

38

socioeconomic groups reported more obstacles than persons in higher socioeconomic
groups. Johnstone and Rivera also described continuing education as "clearly a middle-
and middle-upper class phenomenon" based on that research, and that persons in lower
socioeconomic levels reported less interest in education, even when they knew it could

elevate job or income status.

Johnstone and Rivera defined an educational activity as any organized learning experience
that 1) provided an opportunity for the learner to acquire some type of knowledge,
information, or skill, whether informational, recreational, or job related, and 2) was
organized around some form of instruction. They considered only organized forms of
self-study, however reading was included when it was part of a planned study. They also
pointed out that this definition of an educational activity is much broader than that defined
by the national Oflice of Education (OE) in a related survey of adult education. In the OE
study, self-instruction such as correspondence courses, on-the-job training, private
lessons, television courses and other home-study activities were excluded (Johnstone and

Rivera, 1965).

As Johnstone and Rivera (1965) examined deterrents to participation in educational
activities for adults, they identified two general categories for those individuals who were
deemed at a level of ”high participation readiness”. These two categories included
deterrents which were 1) external to the potential participant, i.e, beyond the control of

the individual, and 2) internal, i.e., within the individual's control such as the participant's

39

attitude or disposition toward participating. The investigators provided respondents with
10 possible deterrents fiom which to choose. Of these, barriers that tended to be of the

external variety were more often an influence than internal barriers.

Cross (1981) identified three major categories of deterrents for adults regarding
participation in educational activities. These three categories were termed situational,
institutional, and dispositional. She defines situational deterrents as ”those arising fi'om
one's situation in life at a given time”, e.g., lack of time due to job responsibilities, or lack
of money to participate due to income level. Institutional deterrents are those which are
created by the institution ofi‘ering the educational activity, e.g., unattainable requirements
for registration or participation, courses or workshops offered at inconvenient times or
locations, or non-relevant topics. Dispositional deterrents refer to negative, internal
feelings toward participation in educational activities, e. g., lack of confidence to complete
the activity or even to attend, general dislike for studying or a disinterest in continuing
education, a desire to do something else instead of participating in an educational activity,

etc.

In summarizing 30 state and national surveys, Cross (1981) found that situational barriers
led the list in all of the surveys she examined. The cost of education and lack of time led
all other barriers substantially. She suggests that those who have the money do not have
the time, and those who have the time, do not have the money. Individuals in the 25-45

year old age bracket appear to state lack of time more often than other age groups, due to

40

job and home responsibilities. Low income and younger people tend to report lack of
money as a deterrent more often than those having higher incomes or older individuals.
Cross also mentions that women indicate cost as a deterrent to participation more often
than men. One national survey indicated women more often were expected to pay for the
cost of participation fi'om personal funds, while men were more often supported by
employers. This phenomenon was reportedly due to the greater majority of men employed
outside of the home at the time of the study (Cross, 1981). This was thought to be the

major reason why females reported cost as a deterrent more often than men.

Cross cautions that having the ability to pay is not the same as having the willingness to
pay. Also she suggests that the cost of the program is sometimes used as a maneuver by
respondents to hide their real feelings. The phrase ”costs too much", is often a socially
acceptable response that people use when it may really mean that the value of the activity
is not worth the cost, or is not a priority at the time. Therefore, it is sometimes dificult to
measure in a survey whether individuals economically cannot attend (situational), or
whether they just do not see a value derived by attending (dispositional). Cross also states
that dispositions! barriers may be underestimated in the literature, since it is more socially
desirable to state that one is too busy than to say that one is not interested in learning, or
not confident in their learning ability. This is supported by other studies which have
shown that individuals may not indicate lack of interest in learning as a deterrent for

themselves, yet they will cite this as a deterrent for others (Cross, 1981).

41

In the national study mentioned by Cross, institutional barriers included references to the
amount of dedication expected by the learner, e.g., requirements to attend school full time,
or courses meeting more often than the learner could attend. But Cross also mentions
that inadequate information about educational offerings also can deter learners from
participating. Cross groups institutional barriers into five areas, i.e., problems due to 1)
scheduling, 2) location, 3) course relevancy, 4) procedural or time requirements, and 5)
lack of information about the program. She states that "lack of information” about the

program is cited less frequently than other deterrents.

Darkenwald and Merriam (1982) also cite lack of time and cost as the two barriers to
participation most frequently reported by adults. They also agree with Cross (1981) that
individuals will often cite lack of time or cost as a deterrent, rather than give a socially
unacceptable reason such as not being interested or not feeling competent in their ability to
learn. Darkenwald and Merriam added a fourth barrier to those ofi‘ered by Cross (1981),
and expanded the meaning of another. They grouped barriers into categories labeled
instimtioml, situational, informational, and psychosocial. The informational category
was added to reflect the lack of information that is ofien cited by respondents when trying
to locate educational opportunities. They redefined the term “dispositional” as
psychosocial which they report more broadly represents the category of attitudinal and

dispositional responses.

Related to the informational barrier, Darkenwald and Merriam (1982) found that one-third

42

of all adults indicated that they had “no knowledge whatever of educational resources in
their communities”. This compares with the findings of Johnstone and Rivers (1965) who
reported that this is more often true with disadvantaged and low income sectors of the
adult population. Their study showed that about one-fourth of the adult population did
not know where to go or who to ask about educational programs in their communities.
Johnstone and Rivera also reported in one study that 85 percent of individuals of high
socioeconomic status living in middle-sized cities knew of at least one place to go to find
out about educational activities in their community. The same study showed that of
individuals from low socioeconomic status in smaller towns, only 19 percent knew of at
least one source for getting information on educational opportunities. Johnstone and
Rivera (1965) and Darkenwald and Merriam (1982) found many similarities in their

research which was used in subsequent research.

In these earlier studies, the concept of deterrents was examined through more simplified
and intuitive methods (Scanlan and Darkenwald, 1984). Motivational research had not
been successful in predicting participation, although motivational research did identify
forces which motivated persons to learn (Scanlan and Darkenwald, 1984). More recently,
Scanlan (1982) examined adult deterrents to participation in educational programs in New
Jersey with hopes of constructing an instrument that would 1) more completely identify
and explain deterrents to participation, and 2) be useful in predicting participation
behavior in adult audiences. That study focused on health professionals, including

physical therapists, medical technologists, and respiratory therapists who resided in that

43

state. Results of this study showed that a multidimensional construct existed and that
using principal components analysis and multiple regression, the deterrent source factors
which emerged could be used to predict participation behavior. To perform that study, an
instrument was developed entitled the “Deterrents to Participation Scale (DPS)”. A
number of studies were performed in the 1980's which tested the DPS instrument and
modified forms of the DPS. Scanlan (1982), Darkenwald and Valentine (1985),
Weischadle (1988), Davis ( 1988) and Martindale and Drake (1989) used the DPS
methodology to investigate barriers to participation in educational activities for various
audiences. Using Orthogonal Rotation, deterrent factors were identified in each of these
studies. Type, number, and importance of the deterrents varied with the target population

(Table 1).

In the Scanlan study, a 6-factor solution of deterrent source variables was identified
which accounted for 54 percent of the scale variance (Scanlan and Darkenwald, 1984).
These factors were termed Disengagement, Lack of Quality, Family Constraints, Cost,
Lack of Benefit, and Work Constraints. The predictive utility of these factors was
examined through a regression equation using participation-nonparticipation as the
dependent variable. With the exception of Factor 6, the source variables were useful in
predicting participation behavior, explaining 41 percent of the variance (Scanlan and

Darkenwald, 1984).

 

 

 

 

 

 

 

Eu:— 3:90—3—
rmaasfi 23.2 been 8858c aspasu 88E80 280 as: and
sea. been no as. .e as 8E. .80 new. sea— e 2853:
3.8.5 888 Sasea amuse 3580 based
.25 see no 9:82 c303 8.; 28o 3:: tan
“SEE essence 83833 33380 as:
.6388. 338.8 moo 532.. 25 cease?
8.553 Ea:
Ease.— aeoE aspESU 8.80 88380 gage
38%.. a8 1820.. 33 8.: co as. s as. gassed
Aves—V
essence «moan assuage £95 .85 eased
e03 coed sou been seen swag e 5.8»
a 3.3..— a Lego-h e .58...— m 38¢..— v 38:...— n .83.: a Stab — .38...— been»:

 

 

 

 

 

 

 

 

 

seem we: eat; a ass..— .s 382.55 3.2:... ._ sec.

 

45

In 1985, a general study on New Jersey households was performed using a new prototype
version of the DPS instrument coined the “DPS-G” (Darkenwald and Valentine, 1985).
The new instrument which explored deterrents to participation in educational programs
among the general population also led to a six-factor solution of source variables although
only one source variable (Cost) emerged that was also identified in the original DPS study.
In order of explained variance, the six DPS-G factors were termed Lack of Confidence,
Lack of Course Relevance, Time Constraints, Low Personal Priority, Cost, and Personal
Problems. This difl‘erence in the factor solution was anticipated due to the differences in
the two populations. This led to recommendations that new populations be studied using
modified forms of the DPS or DPS-G instrument (Darkenwald and Valentine, 1985;

Scanlan and Darkenwald, 1984).

Two more investigations into the deterrent construct were performed in 1988 by doctoral
students at Rutgers University. Weischadle (1988) examined real estate professionals’
response to a combination of the original DPS, an AACES (Adult Attitudes Toward
Continuing Education Scale) instrument, and a Behavioral Index. Her study resulted in
five deterrent source variables surfacing which accounted for 47.5 percent of the scale
variance. These factors were identified as Time Constraints, Program Deficiencies, Cost,
Scheduling, and Incongruent Dispositions. Two of these factors, Cost and Time
Constraints were also seen in the Darkenwald and Valentine DPS-G study. The results
support findings by Johnstone and Rivers (1965) and Cross (1981) in that cost and lack of

time continue to be important deterrents. Weischadle’s study did significantly differentiate

46

between the deterrent factors for real estate professionals, and supported the construct
that deterrents are multidimensional in nature. Her study also demonstrated that real
estate professionals may be more inclined to participate in educational activities that are
work-related based on survey response. Using multiple regression in relating the DPS
scores, attitude scores, and participation behavior, Weischadle did report the ability to

predict participation behavior.

The second study that year by Davis (1988) examined deterrents to participation as
reported by bank managers. Davis used a modified form of the DPS instrument referred
to as DPS-B. Seven factors were identified and labeled Course Deficiency, Time
Constraints, Lack of Benefit, Negative Disposition, Cost, Other Options, and Other-
Dependents. The factors entitled Time Constraints, Lack of Benefit, and Cost were
reported in the earlier DPS studies. Davis also incorporated role-related variables such as
bank size, bank type, job level, etc. Support was documented regarding the deterrent
construct being multidimensional. Predictive measurements were significant, however
Davis indicated that explained variance was small for the deterrent factors in relation to
voluntary work-related participation levels. Therefore he states that the study, in itself,

does not prove to be useful as predictor for participation behavior at a practical level.

Another study was performed by Martindale and Drake (1989) with Air Force Personnel
where eight deterrent factors were identified. Of these source variables, four had been

identified with previous audiences, i.e., Lack of Course Relevance, Lack of Confidence,

47

Cost, and Time Constraints. The remaining four source variables, i.e., Lack of
Convenience, Lack of Interest, Family Problems, and Lack of Encouragement were new
factor labels. The results of this study also showed that institutional and situational
deterrent items ranked highest in mean score while dispositional items ranked lowest.
Martindale and Drake did not test for predictability of the deterrents with regard to
participation behavior. As found in earlier research, lack of confidence was found to
increase with age and decrease with educational attainment. Also, younger Air Force

personnel were less deterred by lack of confidence but more deterred by cost.

In 1994, the DPS methodology was used to investigate participation in continuing
professional education programs by university-based nurses at Syracuse University
(Essman, 1994) The study incorporated the original DPS instrument, along with a nine-
item Continuing Leaming Activity Survey and inventory of sociodemographic and clinical
practice variables. The study revealed five deterrent factors identified as Alternative
Choices, Program Constraints, Cost/Benefit Constraints, Time Constraints, and Family
Constraints. Findings of the study supported the construct in being complex and
multidimensional, however the results were not useful in predicting the participation

behavior of university-based nurses.

There is repeated evidence that the deterrent construct is more complex and
multidimensional than that proposed by earlier investigators, however many variables are

related in much the same way as those based on the earlier works by Johnstone and Rivera

48

(1965), Cross (1981), and Darkenwald and Merriam (1982). A general agreement in the
literature reveals that lack of confidence becomes more of a deterrent as a person ages and
with lower socioeconomic status, but decreases as a deterrent with educational attainment.
Cost appears to be a stronger deterrent for females than for males, and for young age
groups more than older age groups. Low personal priority or lack of interest in
participating in educational activities is shown to be a stronger deterrent with individuals
of lower socioeconomic status, and higher for individuals with higher educational levels.
Time and lack of money repeatedly surface as being two the most commonly reported
barriers. And in studies with a cross section of income levels, age groups, and
socioeconomic levels, institutional and situational barriers continually emerge as more

important deterrents to participation in educational activities than dispositional barriers.

Demographics and Barriers
Previous studies by researchers in adult education have reported significant findings about
the adult population and participation behavior. While Scanlan and Darkenwald (1984)
state that sociodemographic traits only account for about 10 percent of the variance in
studies on participation behavior, it is important to examine the similarities fiom various

studies.

Gender
While Darkenwald and Merriam (1982) saw negligible difl‘erences between participation

and non-participation due to gender, J ohnstone and Rivera (1965) reported that women

49

identified more obstacles to enrollment in courses than men. Also they found that more
women than men declared busy schedules and lack of energy as barriers (Johnstone and
Rivers, 1965). Merriam and Cafl‘arella (1991) state that women on the average paid less
for education than men. Cross (1981) states that white females are the only population
subgroup where the majority of learners are supporting educational costs from their own
resources or fiom family funds. Darkenwald and Merriam (1982) concurred with those
findings stating that educational costs for men were more often supported by employers
while women paid educational costs from personal or family funds. Cross (1981) suggests
that this may be why women are more likely than men to cite cost of attending an
educational activity as a deterrent. She states that men fiom a given socioeconomic status
(SES) are willing to pay more for education than women of the same SES. Weischadle
(1988) found that females are more likely to cite cost and scheduling constraints as a
deterrent than men. Findings by Davis (1988) in a study with bank managers and
Darkenwald and Valentine (1985) in a study of the general public also showed that
females more often cited cost as a deterrent than did men. Davis (1988) also reported that
course deficiencies and the lack of benefit fi'om attending were more of a deterrent for
male bankers than female bankers. Darkenwald and Valentine (1985) in studying general
audiences found males to indicate lower personal priority in participating in educational
programs than women, while women cited personal problems such as child care, family

concerns, etc, as more important deterrents than did men.

50
Age
Johnstone and Rivera (1965) stated that older adults reported more obstacles than
younger adults. They also found that younger adults cited financial restraints more often
than older adults, while older adults appeared to lack more confidence. Some older adults
also stated that they felt they were ”too old to learn”, however differentiation between
young and old was more pronounced in the lower socioeconomic levels. Cross (1981)
states that young adults are more likely to see cost as a deterrent than are older adults.
She comments that young people tend to see education as a "right”, i.e., something that
should be provided, where older people see education as a ”privilege”, and thus are more
willing to pay. Cross (1981) also found that older people on fixed incomes are more
willing to pay than younger people, and lack of time appears to be more of a problem for
people in their 30's and 40's, than for the younger or older. Older individuals also tend to
have more problems with transportation (Cross, 1981). Weischadle (1988) saw time
constraints decrease with age for real estate professionals but cost, unlike the findings in
most studies, became more important as a deterrent with age, . Davis (1988) reported
that older bank managers were more likely to cite course deficiency or negative
dispositions toward participation, while younger bankers appeared to be more restrained
by time and money. Darkenwald and Valentine (1985) observed that lack of confidence,

cost, and personal problems declined with age as deterrents to participation.

51

Level of Education

Houle (1961) stated that "the most universally important factor [regarding participation in
adult education] is schooling”. He states that the higher the level of education, the more
likely the individual is to take part in some form of continuing education. Cross (1981)
stated that the highly educated proclaim lack of time as a barrier more than those with
lower levels of education. Weischadle (198 8) found that real estate professionals’
dispositions toward education decreased as the level of educational attainment increased.
Darkenwald and Valentine (1985) in working with general audiences saw that lack of

confidence, cost and personal problems declined as level of educational increased.

Socioeconomic Status

Johnstone and Rivers (1965) reported that persons in lower socioeconomic status (SES)
groups reported more obstacles than persons in higher socioeconomic levels. They also
identified four tendencies: 1) Persons in the lower SES groups more often reported
financial restraints as a barrier to participation than did persons in higher SES groups, 2)
Persons in the lower SES groups were more likely to cite the lack of facilities in their area
as a barrier, 3) Persons with a high SES say they are too busy to participate, 4) Persons
with a middle level SES cited more difliculty in getting away fi'om home in the evenings to
participate. Johnstone and Rivera also found that persons of low SES were more likely to
have not heard of any courses interesting enough to attract them into enrolling. Persons
with high income levels tend to indicate lack of time as a barrier, more than persons with a

lower income. Darkenwald and Merriam (1982) found that individuals with low SES

52

demonstrated less interest in education even though they were aware that educational
attainment could lead to better jobs and higher income levels. Weischadle (1988) reported
that as income increased, real estate professionals became more concerned about program
deficiencies. Darkenwald and Valentine (1985) stated that as income for general adults

increased, cost and lack of confidence became less of a deterrent.

For the most part, the findings in the above studies agree and are logical. In summarizing,
females cite cost as a deterrent more so than males, however males appear to be more
concerned about program relevancy, and state that low personal priority is a stronger
deterrent for participating than it is for females. Older individuals often state lack of
confidence as a deterrent to participation in learning experiences. Although studies have
shown mixed results regarding time constraints and cost, it appears that typically an older
person has more money and less time to attend educational activities. It has repeatedly
been shown in the literature that as level of education increases, lack of confidence
decreases and personal priority for participation increases. Socioeconomic status (SES) is
also fairly dependable in dictating participation in educational activities. Generally, the
higher the SES, the more inclined an adult is to participate, costs are less of a deterrent,
and time constraints become more of a barrier. On the other hand, persons with a low
SES tend to have more time available, less money, and a lower personal priority toward

participation in educational activities.

53

Summary of Adult Education And Deterrents
The literature has shown that most adults prefer to participate in some type of continuing
education activities, although the amount of participation and the types of deterrents vary
with sociodemographic group. Motivational research, although comprehensive in
demonstrating the reasons why individuals seek to learn, has not shown the ability to
predict participation for adults according to Scanlan and Darkenwald (1984). Earlier
research focusing on the barriers to participation provided logical reasons why people do
not participate but the construct was not examined as a multidimensional network of
relationships (Scanlan and Darkenwald, 1984). The deterrent to participation scale, and
modified forms of the scale, provided a new pathway to investigate the deterrent construct
and have been able to provide a prediction of participation behavior in certain studies
(Scanlan, 1982; Weischadle, 1988). Studies have repeatedly shown that barriers to
participation do not show up as a single deterrent type, but rather appear as a number of
deterrents working together to deter an individual fiom participation. Previous
assumptions regarding sociodemographic variables and deterrents found in early forms of

adult education research have been reconfirmed through DPS-type studies.

Hobbs (1992) states that nrral communities have lagged behind in the proportion of
college graduates and number of jobs requiring higher levels of education. Those who
attain post-secondary degrees then leave the rural community for better jobs. Land use
trends may be changing this balance, as individuals with higher socioeconomic status move

their homes to rural areas. Hobbs states that there is a need for contimring education in

54

rural communities as regional and national changes impact rural communities. Farms are
growing in size and becoming more specialized (Bemsten, 1993). Technology and
cormnunications are becoming increasingly complex yet probably offer the greatest return
on investment (Ferris, 1992). At the same time, experts state that numbers of on-fann
laborers are decreasing in number and that farm owners and managers will depend on
technologies to replace that labor and increase efliciency (Shapley, 1992). To compete
and survive in a global market, farmers will have to continue to participate, if not increase

participation, in educational activities or be driven out of business by those who do.

As a result, the public and private organizations that provide educational opportunities for
agricultural producers must be ready. They must know why farmers attend and what
deters them from attending. As stated by Merriam and Darkenwald (1982), “... in adult
education, the efl‘ectiveness and often the survival of educational programs depends on a
thorough understanding of the needs, problems, attitudes, and preferences of its clienteles

and potential clienteles.”

Return on investment to educational institutions who work with producers will be as
important as it is to the producers themselves, and registration fees may have to be
included as a way to cover implementation costs. This is especially true for public entities
which have limited personnel and operating budgets. While recognizing that the
identification of farmers’ deterrents to participation will greatly aid those organizations in

allowing or encouraging farmers to participate, the ability to predict participation behavior

55

will be an even greater tool in the agricultural educational arena. The current study on the
“Deterrents to Participation in Educational Activities by Michigan Farmers” will attempt
to explore and unveil some of these unknowns, and provide useable data for all educators

working with the agricultural industry.

CHAPTER III

METHODOLOGY
Introduction
This section of the dissertation presents the methodology that was used in the study.
Topics discussed include the nature of the study, the research design, a review of the
research questions, dependent and independent variables, internal and external validity,
generalizations to populations, reducing sampling bias, instrument development, data

collection, correction for non-response error, and data analysis.

This study, performed via a mail survey, takes an applied approach to the problem

addressed, and falls into the category of educational research. Ary et. al. (1990) state:

”Educational research is the way in which one acquires dependable and
useful information about the educative process. Its goal is to discover
general principles or interpretations of behavior that can be used to explain,
predict, and control events in educational situations-in other words,
scientific theory. "

56

57

Nature of The Study
In this study, Michigan farmers were surveyed via a mail questionnaire (Dillman, 1978), to
examine aspects of the deterrents construct. The study attempts to 1) develop a valid and
reliable “deterrents to participation” survey instrument for farmers, 2) identify significant
factors that serve as deterrents to farmers' participation in educational programs, 3)
observe whether difl‘erences exist in the deterrents identified based on socio-demographic
difl‘erences in the sample population, 4) determine whether the cost of attending an
educatioml program serves as the most important deterrent to participation for farmers, 5)
observe whether there is a statistically significant relationship between the type and
importance of deterrents identified by participants and non-participants that would allow
prediction of participation behavior? A goal was established to develop and test a model

the would predict participation behavior by farmers.

To date, much research has been performed on learning style, learning orientations, and
those stimuli which motivate adult participation in educational activities. However, based
on the literature, motivational research has not been substantially useful in predicting adult
participation behavior (Scanlan and Darkenwald, 1989). In addition, limited research has
been performed in the area of deterrents to participation, and this is virtually non-existent

in regard to farmers. These findings and sources have been reviewed in Chapters I and II.

58

Research Design
Introduction
The nature of this study required that a descriptive research design be used. Descriptive
research is recommended when a population will be examined and the observations
organized, sununarized and described (Ary et. al., 1990). In this study, farmers were
questioned via a mail survey as to the barriers that deter them from participating in
educational activities. The term ”educational activity” was defined as any type of
organized, group learning experience, whether recreational, informational, or
work-related. Source variables (”factors") were identified and compared using factor
analysis.

Inferential statistics were used to assist in predicting the results of this study to the general
population of farmers in Michigan. Responses were analyzed using Orthogonal Rotation,
Pearson Product-Moment Correlation, Friedman’s Test, Repeated Measures Test,
Logistical Regression, and One-Way Analysis of Variance. Factors identified were
examined with regard to sociodemographic groups and with regard to participants and
non-participants. In this study, farmers were termed "participants" if they had participated
in at least one organized, educational activity in the past twelve months prior to receiving

the questionnaire.

Sociodemographic groups were compared based on gender, age, level of education,

amount of annual gross farm sales (farm income), type of farming enterprise, farm

59

employment status (i.e., part-time or full-time), and whether they also held an off-farm job
in addition to farming. Statistical analysis was performed to identify deterrent factors and
to see if there were relationships or differences between these groups regarding the factors

that influence their decisions not to participate in educational programs.

Instrument Development

The scale instrument was modeled after the Deterrent to Participation Scale for General
Audiences (DPS-G), which was developed by Darkenwald and Valentine (1985). The
DPS-G was a new prototype fashioned fi'om the original Deterrent to Participation Scale
(DPS) developed and tested by Scanlan (1982). After examining both scales, the DPS-G
appeared more easily adaptable to a farming audience. The new instrument, modified for

farmers, would be referred to as the Deterrent to Participation Scale for Farmers (DPS-F).

A focus group process was used to identify the items to be asked in the survey and to
enhance pre-testing face validity (Krueger, 1994; Heinze, 1996). Focus group meetings
consisted of two groups of farmers from two Michigan counties. The first focus group
meeting was held in Eaton County, Michigan and was comprised of Baton County
farmers, while the second meeting was held in Ingham County, Michigan with Ingham
County farmers. While Krueger (1994) suggests involving a minimum of 4 persons in the
focus group, only 3 of the 9 farmers invited attended the first focus group meeting. Five
of the 12 invited farmers attended the second meeting. These meetings were held during

seasons when farmers are not busy, i.e., early summer and early autumn, respectively.

60

The format for focus group questioning followed that suggested by Krueger (1994).

Focus group participants were asked to indicate barriers to their participation in any
educational activities, regardless of whether the activities that were farm-related,
recreational, or just informational. “Educational activity”, for purposes of the study, was
explained to each focus group as a non-formal, organized educational activity. It excluded
such things as reading and self-study, college or formal course work, etc, and included
such activities as farm tours, workshops, weed control meetings, golf lessons, Bible study,
etc. The resulting responses were used to assist in constructing the research

questionnaire.

Cross (1981) cautions researchers regarding respondents who may lie in order to provide
a socially acceptable answer. Krueger (1994), supporting this belief, states that ”people
often want to tell us how they wish to be seen as opposed to how they are." In other
words, it is possible that a farmer may not list socially, non-acceptable answers, such as "I
am not interested in learning” or answers that may negatively impact his self-esteem, e.g.,
”I do not feel confident in my ability to keep up with the class”. Therefore, in order to
better identify these types of items for the questionnaire, participants in the focus groups
were not only asked to identify valid deterrents for themselves, but they were also asked
to identify the deterrents they thought "other" farmers might name. The investigator of
this study also has a strong agricultural background, which was valuable in interpreting
farmer input regarding perceived deterrents. However, participants were still asked to

clarify any comments that were unclear to the proctor or group, or that could be

61

interpreted in more than one way.

Using the information compiled fi'om the focus group sessions, questions were generated
for the pilot survey. Deterrent items from the original DPS-G instrument that were also
mentioned in the farmer focus groups were left intact for the DPS-F instrument. Those
deterrent items in the DPS instruments that were not specifically mentioned in farmer
focus group sessions and did not seem to apply to most farmers, were either merged into a
related item that surfaced during the focus group sessions or were deleted from the
questionnaire. For example the item, “Because of transportation problems. . .”, does not
apply well to farmers since farmers must have vehicles for moving from field to field, or
traveling to town for parts, groceries, etc. Most farmers have at least a pick-up truck for
hauling supplies, etc, or in some cases a horse—drawn vehicle or van in the case of certain
cultural groups, such as Amish and Mennonite families. Therefore, this particular item
was not included in the DPS-F questionnaire. Deterrent items related to location of the
program allowed respondents to provide “location” as an alternative deterrent but did not

focus on the method of transportation being the cause if this was a deterrent for them.

A total of 56 deterrent items surfaced at the Eaton County focus group session, while 39
deterrent items were named at the Ingham County session. These two sets of deterrent
items were compared with each other and with the DPS-G instrument. After merging
duplicate or triplicate deterrent items and eliminating DPS-G items not related to the

occupation of farming, a total of 39 deterrent items remained. These 39 items were

62

incorporated into the DPS-F instrument as Likert scale items. This number was similar in

size and scope to other DPS scales that have been used.

Prior to testing the pilot questionnaire, a group of qualified university faculty and
Extension staff members were asked to review the instrument for face validity (Dillman,
1978). Suggestions were made to change the wording on an open-ended question near
the end of the survey, and to drop the labels, “Part I, Part II, etc” from the various
sections of the questionnaire. Typographical errors were discovered and corrected. A
suggestion was also made to place sociodemographic questions at the end of the survey,
especially since one question asked respondents to indicate their income, i.e., annual farm
sales. Although this was an optional question, it 'was agreed that respondents would more
likely complete all or most sociodemographic questions, once they had invested time in the

Likert scale portion of the questionnaire.

As a second step in testing for face validity, 5 individuals who work directly with farmers,
e.g., Extension agents, USDA Natural Resources Conservation Service stafi‘, and
agribusiness representatives were also asked to review the questionnaire to help ensure
that it would be readable and acceptable to farmers (Dillman, 1978). Several suggestions
were made, some of which did not follow Dilhnan’s guidance. Those suggestions which
did not follow established mail survey procedures nor comply with Dilhnan’s
recommendations were disregarded. The first question regarding the number of activities

attended was reworded slightly based on this phase of the input.

63

After validity was assumed, a pilot test was administered to 11 Michigan farmers,
comprised of 9 males and 2 females (Dillman, 197 8). The pilot group was comprised of
Ingham County farmers who were not involved in focus groups nor individual input to
construct the scale items. Potential pilot participants were invited by telephone and these
consisted of individuals known to participate in educational programs and not known to
participate. During the verbal invitation, pilot participants were told that they would be
completing a pilot survey as part of a research project, but no explanation, descriptions,
etc, were provided. On the day of the pilot testing, the proctor welcomed the group,
thanked them for participating, and stated that they he would explain the reason for the
pilot survey after they completed the exercise. This method was implemented to better
assimilate a person receiving the cover letter and questionnaire in the mail. Pilot
participants were asked to read the cover letter and then complete the questionnaire
without the proctor’s assistance. Participants were asked to place an ‘x” or “check-mark”
next to any item that was confirsing or of concern. These items were brought before the
entire group for discussion and potential concurrence among the members after taking the
pilot test. Again, to assimilate a person completing the questionnaire as a mail survey, no
time limit was set. When everyone had completed the questionnaire, the proctor asked
everyone to put their pencils down and make no more marks on the questionnaire.
Participants were then ofi‘ered an opportunity to provide verbal feedback, however they
were again instructed not to change any answers or comments on their questionnaire.
Many participants offered constructive comments. In the first item of the questionnaire,

there was agreement that farmers should not be asked to name the educational activity(s)

64

that they had attended, therefore that portion of the first section was omitted from the final
questionnaire. Other suggestions for rewording questions were made, although
participants all stated that the questionnaire was generally clear and concise. Therefore, a
decision was made not to reword those questions unless the reliability analysis identified a
problem Results of the pilot test were analyzed for reliability using Cronbach's Alpha
Coeficient with alpha = .8831 for the 39 items. Although five items showed a slightly
higher alpha value if deleted (optimum alpha value of .8901), four of these items appeared
in the original DPS-G and the remaining item was specific to the farming audience, i.e.,
“Because the activity schedule conflicted with my ofl'-farm jo ”. All five items had been
developed or reaffirmed through focus group input and therefore, these items were
retained. No additional changes were made to the questionnaire after the reliability
analysis, and that version of the questionnaire was considered the final draft of the

instrument.

Research Instrument Summary

Using the input fiom farmer focus groups and one-on-one input, a modified scale was
developed for use with farmers. This scale was based on the Deterrents to Participation
Scale for General Audiences (DPS-G) developed by Darkenwald and Valentine (1985),
with consideration given to the Deterrents to Participation Scale (DPS) developed by
Scanlan (1982), and the Deterrents to Participation Scale for Bankers (DPS-B) developed
by Davis (1988). This new modified scale, which is referred to as the Deterrents to

Participation Scale for Farmers (DPS-F), attempts to identify those factors which serve as

65

deterrents to farmers' participation in educational activities.

The DPS-F consisted of a five-item Likert Scale mail questionnaire. In addition to the
Likert Scale format, one question was asked regarding participation in educational
programs during the 12—month period prior to receiving the questionnaire. As with other
DPS studies, those respondents who indicated any type of participation in one or more
organized, educational activities during the 12 month period prior to the study were
considered a "participant”. Those who had not participated in any organized, educational
activities in the 12-month period prior to receiving the questionnaire were considered a
”non- participant". According to Tough (1971), the average adult normally participates in
eight learning projects per year. Cross (1981) states that about one-third of the adult
population participates in some type of ”organized learning" activity each year. Therefore,
for possible future use with the data, respondents were allowed to select one of four
options, i.e., that the respondent had participated in no educational activities (“none”), 1-5
educational activities, 6-10 activities, or 11 or more activities in the past year. These
categories were selected in units of five, i.e, “1-5", “6-10", etc, to help reduce confirsion
for growers who may not remember the exact number of activities they had attended.
These responses were later utilized during the analysis to see if there was a relationship

between the deterrent factors and participation behavior.

In addition, a section was provided to determine socio-demographic characteristics of the

sample population. These questions helped to determine differences due to gender, age

66

group, level of education, type of farming enterprise, amount of gross farm sales, farming

status, i.e., firll- time or part-time farmer, and those with or without ofilfarm jobs.

In order to achieve the highest level of response in the mail survey, questionnaires were
mailed during the months of March and April 1997 . These are slower months for farmers
in regard to farming activities, and should have allowed them time to complete and return
the survey more readily before becoming busy with spring field activities. A copy of the

research instrument is included as Appendix B at the back of this dissertation.

Review of The Research Questions
This section provides an overview of the research questions. Table 2 identifies the
research questions that were asked, and those statistical tests that were applied. In

addition, questionnaire items that were critical to the analysis are identified.

67

Table 2. Research Questions, Tests, and Related Items

 

 

 

 

 

 

RESEARCH QUESTIONS TESTS RELATED ITEMS
1. Using a farmer-oriented form of Means, Frequencies, T- Deterrent Items::
the Deterrents to Participation Tests, Cronbach’s Alpha #1 - 39.
Scale (DPS-F), can significant and Coeficient, Factor
distinct source variables or factors Analysis with Orthogonal
be identified that deter adult Rotation using the
farmers from participating in Varimax procedure with
organized adult education Kaiser Criterion,
activities? Repeated Measures Test.
2. Is there a difference in the type Means, Pearson Preduct- Deterrent Items:
and/er importance of deterrent Moment Correlation #1 - 39;
factors for farmers based on the Ceeflicient, One-Way Sociodemographic
socie-demegraphic variables of Analysis of Variance with Items: #1,2, 5, 6, 7,8,
gender, age, level of education, Tukey’s HSD pest hec 9.
type of enterprise, arneunt of gross tests.
farm sales, farming status and on-
or ofllfarm employment?
3. Is “cost” the most statistically Friedman’s Nen- Deterrent Items:
important factor that deters Parametric Test for K- #10,#22,#37
farmers fiem participating in Related Samples. (based on results of
educational activities? factor analysis, i.e.,

F actor #6 - Cost);
Factors #1 -5.

4. Is there a statistically significant Pearson Product- First Item on Number
relationship between the type and Moment Correlation of Activities (grouped
irnpertance of deterrents identified Ceeficient, Logistical by participant or non-
by participants and non- Regression. participant) Pg 3;
participants in educational Deterrent Items:
activities that would allow #1 - 39.
prediction of participation

behavior?

 

 

 

 

68

Generalizations to Populations
Generalization to Michigan farmers was addressed through a proportional stratified
random sample of finners (Ary, et. al., 1990). Names and addresses were selected via a
computer process for the Suldy from mailing lists maintained by the Michigan Agricultural
Statistics Service. This division of the Michigan Department of Agriculture maintains a
list of firmers in Michigan, which is comprehensive and represents all types and sizes of
Michigan firms. According to Pscedna (1996), the MASS mailing list includes about 70
percent of Michigan farmers with firm sales over $1000, and is the most complete and
unbiased list available in Michigan. Unlike mailing lists available from CBS or the USDA
Farm Service Agency, the MASS list is not biased toward those farmers who normally
participate in educational programs nor toward those who participate in government farm
programs, i.e., typically larger firms. Names on the MASS list are generated from
combined sources which include, but are not limited to, the USDA Farm Service Agency,
agricultural commodity groups and associations, state government, Michigan State
University Extension, and direct reports to MASS from farmers when farms are sold to
new owners (Pscedna, 1996).

The proportional breakdown of the sample was based on the percent breakdown of
primary enterprise types (by percentage of total) on the MASS list and represents as
closely as possible the percentages of firm types in Michigan. Since the sample represents
the general firm population in the state, the results of this study can be generalized to that

population, assuming a similar response pattern. According to Pscedna, the remaining 30

69

percent of firms not included in the MASS list are typically very small firms. These are
more likely to be missed in the development of the total MASS list, merely due to the
number of very small firms and hobby firms in Michigan (Pscedna, 1996). Also,
according to Pscedna, many of these Michigan firms missing from the list are so small
that the owners do not consider themselves true farmers even though they meet minimum
qualifications as described in the US Census of Agriculture (1992). A small tendency for
error could occur in generalizing to farmers operating these very small firms or hobby
firms. Also, while gender did show acceptable confidence levels in non-response error
tests, the significance level fell exactly at p = .05. There is a slight possibility for error in

generalizing the data based on gender.

Because of the variation in geography, climate, and cultural difi’erences in the United
States, the results of this Michigan study cannot be generalized to the firming population

of the entire country.

Internal and External Validity
Internal Validity
Although this study falls in the category of descriptive research, there are aspects of
internal validity which were considered such as selection bias, instrumentation, testing,
attitude of subjects, and survey irnplernentatien (F raenkel and Wallen, 1993). Where

possible, precautions were taken against such error as explained below.

7O

Selection bias may skew test results if subjects in the group differ on such variables as age,
gender, ability, socie-economic status, and background. The large number in the sampling
group should have helped to bufi‘er this type of error. In addition, a number of key
difi‘erences between members of the audience, e.g., age, income level, etc, were
intentionally difi’erentiated in order to see if there, in fict, is a difference based on these

types of traits.

Instrumentation, i.e., the way in which the survey instrument is developed and how data
are collected, can lead to internal errors of validity. By following Dillman's guide in
instrument development, an attempt was made to minimize errors in instrument
development. Also, since a mail survey was used, errors in data collection bias, i.e., how

questions are asked and scored, were uniferrn.

The term testing refers to the process of actually taking the test or filling out the survey.
With a mail type survey, there is no way to protect against the participant discussing the
questions with a fiiend or relative, thereby influencing how the participant answered the
questions. Some possibility for error does exist here, as in most mail type surveys.

However, again, the size of the sample has helped to buffer any testing efi’ect.

The way in which the participants viewed the study, i.e., the attitude of subjects, may be
the most likely threat to validity in this study. This threat arises when the participant

knows they are part of a study and their answers are influenced due to that knowledge. In

71

this particular study, it is possible that individuals who may not have a desire for learning,
may in fict have answered questions as though they did have that desire. This study
examines a person’s attitude with respect to continuing education, and the survey was sent
fi'om an educational institution (Michigan State University). A possibility for error in
validity does exist since a respondent may not have wanted to state that they are adverse

to continuing education when they are aware that university staff may see their response.

External Validity
Elliot (1990) identifies three types of external validity to be protected, i.e., population

external validity, ecological external validity, and external validity of operations.

Population external validity relates to the sample population being representative of the
population to which the findings will be generalized. The sample constructed for the mail
survey was a proportional stratified random sample (Ary, et. al., 1990) of 70 percent of
Michigan firrners producing over $1000 in firm receipts. Pscedna (1996) states that this
list includes the majority of the farmers in Michigan, and that those not included on the list
are likely to be very small firms. Therefore, population external validity should be high
and generalization to the entire population of Michigan firmers should be acceptable (Ary
et. al., 1990). Ary et. al, also state that fictorial designs strengthen population external

validity regarding interaction of subject characteristics and treatment.

Ecological external validity relates to the physical and mental environment in which the

72

respondent completes the questionnaire and whether this may affect the way in which the
participant responds. Efi‘ects due to pre-test or post-test sensitization and interaction
between history and treatment effects are eliminated since only one sampling was
incorporated, i.e., no pre- and post-tests were applied. Pretest sensitization to those
individuals selected for piloting the questionnaire for reliability was not a threat since those

same individuals were not selected to participate in the actual survey.

Novelty and disruption effects were avoided or reduced in this study, since questionnaires
were completed at the finners residence and questions were based on the firmer’s past

experience or attitudes rather than an artificial treatment.

Ecological external validity was strengthened due to the use of a mail survey, since there is
no environmental setting or classroom efi’ect that is unique to one particular group.
Differences in response due to the age of the respondent, level of education, etc, were
intentionally examined as a part of the study. A weakness in ecological external validity
could appear in reactive afi‘ect, sometimes referred to as the Hawthorne Efi‘ect, where
respondents react difi‘erently when they are aware of being part of a study. Ary et. al.
(1990) state that this efi‘ect can cause respondents to alter their responses as a ”good
subject" rather than to display their typical behavior. Krueger (1994) and Babbie (1983)
also state that this type of study can lead to respondents giving socially acceptable
answers. Respondents were encouraged (via the instructions in the questionnaire) to be

honest in answering the questions and were told that their individual responses were

73

anonymous and would be kept confidential.

External validity of operations is threatened when one experimenter, interrogator, or
evaluator has a different effect on the respondent, for example when one evaluator is more
enthusiastic than another in the course of implementing a telephone survey or interview.
This is sometimes referred to as the "experimenter effect". The use of a mail survey to all
members of the sample, eliminated this threat. The telephone non-response process was
conducted by 6-8 professional interviewers of MASS on a week night between the hours
of 4:30 pm and 8:30 pm. After completing this telephone non-response error check, it
was discovered that selected male respondents were likely still doing farm work, as a high
percentage of women were answering the telephone. Therefore, the telephone survey was
not used to protect for non-response error. There was no possibility for external validity
of operations due to the telephone survey. Non-response error was instead tested using a

comparison between early and late respondents (Miller and Smith, 1983; Elliot, 1990).

Operational definitions of the independent and dependent variables are a threat, since
difi‘erent farmers could have interpreted deterrents items or ”educational activities” in a
different way than intended in the questionnaire. Emphasis was placed on keeping
deterrent items clear and concise, and in instructing the respondents as to a single meaning
for the term “educational activities”. Face validity activities and pre-test reliability

analysis should have helped to reduce this threat.

74
Reducing Sampling Bias
Sampling bias is addressed below in categories of sampling error, frame error, selection,

measurement, and non-response error.

Sampling Error

Sampling error is the difi‘erence between the characteristics of a sample and the
characteristics of the population fi'om which the sample is drawn (Elliot, 1990). Since the
sample was taken via a proportional stratified sample of the MASS list, the possibility of a
sampling error is greatly reduced (Elliot, 1990; Ary et. al., 1960). Stratification of the
sample provided assurance that subgroups such as equine producers or fiuit growers were
represented in the sample in proportion to their numbers in the Michigan population.
Randomization allowed all members of these subgroups, regarding age, gender, amount of
gross firm sales, etc, to be selected. Sampling error is also a function of the sample size,
with the error being larger when the sample is small. In this study, the sample size was
considerably larger than recommended in the literature to achieve significance. Research
shows that with a total available population of 37,000 farmers in Michigan, a sample of
380 farmers should be drawn, in order to provide a 95 percent confidence level (Krueger
and Suvedi, 1996). However, a sample of 1207 firmers was taken from the total MASS
list of 37,000 farmers with 466 useable questionnaires returned. This response provided
suficient numbers for the proportional stratified sample and the analyses based on the
entire group response. Where tests were applied to sociodemographic groups, those with

less than 30 respondents per group were not used to test the hypotheses.

75

Frame Error

Frame error, according to Elliot (1990), is ”the discrepancy between the intended target
population and the actual population from which the sample is drawn". The sample was
intended to represent a cross-section of Michigan firmers. Frame error was reduced by
using the MASS list, which is comprised of only Michigan firrners and is up—dated on a
regular basis. The MASS list represents approximately 70 percent of Michigan finners
earning over $1000 in annual firm sales (Pscedna, 1996). This list is thought to be the
most comprehensive list available in Michigan that includes firmers of all ages, all size
firms, all levels of education, all levels of income, all types of primary enterprises, etc,
and is not limited to those enrolled in government programs or on educational mailing
lists. A tendency for bias in frame error could have occurred regarding very small firms
or hobby firms, since these are the types most likely missed by MASS in developing their
mailing list according to Pscedna, (1996). The Michigan Agricultural Statistics Service

(MASS) provided address labels for the 1207 intended recipients identified in the sample.

As expected, some questionnaires were returned with no forwarding address. These
questionnaires were returned to MASS to check for errors or up-dated mailing addresses,
and then re-mailed. In total, 94 recipients could not be located, were no longer farming,
or were deceased, thereby lowering the accessible sample population to 1113 potential

respondents.

76
Selection Error
Selection bias was reduced through a proportional stratified sample, which provided all
firmers on the MASS list an equal chance within their enterprise type of being selected
(Ary et al, 1999). The sample was computer generated by MASS, which eliminated any
chance of human intervention or bias. The random sample was taken from the entire
MASS list, excluding those who were selected to complete the pilot questionnaire. The
computer was programmed to select farmers on the list only once each. This guarded
against any firrners having more than one opportunity to be selected due to managing

more than one firming enterprise, e.g., dairy and field crops.

During the telephone survey for non-response error check, a significantly higher response
by females was noted. This was likely due to the timing of the calls, which began at
approximately 4:30 pm in the afternoon and continued until approximately 8:30 pm. It
was not taken into consideration during the plarming phase that those members of the firm
fimily that perform the majority of farm work (males) might still be in the barn or field.
Spouses (primarily females), on the other hand, were in the house preparing the evening
dinner, and therefore the person more often answering the telephone. The telephone
survey was taken solely for non-response error testing and dealt only with
sociodemographic questions in the survey. Because this was a key fiame error which
could be eliminated without afi'ecting the study, the telephone survey for non-response
error check was not used and non-response error was checked using the “early-late”

respondent technique (Miller and Smith, 1983; Elliot, 1990).

77

Measurement Error

Elliot (1990) describes measurement error as any " systematic efiects which operate to bias
recorded results. . .". This includes error due to misinterpretation of questions by the
respondent, instructions for the respondent which are not clear, tendency for the

respondent to give a socially-correct answer, or tendency for the respondent to lie.

Measurement error was addressed through validity and reliability tests (Dillman, 1978).
To do this, a panel of MSU faculty who are fimiliar with testing were asked to review the
questions for clarity and specificity. Next, agriculturalists who were knowledgeable of the
subject matter area and who have experience working directly with farmers were asked to
review the questionnaire. Last, the questionnaire was piloted with a panel of firmers to
check for reliability using Cronbach's Alpha Coeficient. Questions which appeared to be
confusing, ambiguous, or otherwise unusable were deleted or revised. An opportunity did
exist for firmers to provide socially acceptable answers or to lie. However, to reduce this
possibility for error, the survey instructions to the respondent 1) asked for honest answers,
2) stated that the respondent should not place their name on the questionnaire so as to
guarantee anonymity, and 3) stated that individual survey results would be kept

confidential.

Non-response Error
Non-response error was examined for any variations due to those who may not return the

survey. Non-response error was tested using the early-late respondent technique (Miller

78

and Smith, 1983; Elliot, 1990). A threat does exist in that those individuals responding
may be more biased toward the value of education than those not responding. Individuals
not responding to the first item regarding participation may be those who do not normally
participate in educational activities, thus biasing the results of this item. A more detailed

description of the check for non-response error in this study is listed later in this chapter.

Reliability
Instrument reliability was achieved by administering a pilot test of the survey instrument to
a representative sample of the population. Eleven firmers completed the pilot survey. A
reliability test was performed using Cronbach's Coeficient Alpha of Homogeneity to test
for consistency of the survey instrument in measuring results (Ary et. al., 1990). Pre-test
reliability coeficients ranged fi'om .8699 to .8901 with an over-all alpha coeficient of
.8831. Five items were identified that, if deleted, may have slightly raised the alpha
coeficient. After considering the strength of the initial over-all item alpha value along
with close scrutiny of the five items, it was decided to leave the 39 deterrent items intact.
Reliability was again checked after testing based on the 466 returned surveys resulting in

an alpha coeficient of .9012.

Data Collection
Data collection was achieved though a mail survey to a proportional stratified sample of
Michigan firmers, obtained through lists maintained by the Michigan Agricultural

Statistics Service. This follows the prescribed format as stated by Dillrnan (1978) for mail

79
surveys. All target recipients received an identical questionnaire and cover letter, which
were mailed on March 11, 1997. This initial letter was followed by a reminder postcard
seven days later on March 18. A follow-up questionnaire and cover letter were mailed on
April 1, i.e., 14 days after the postcard was sent. In each request, recipients were
instructed to return the completed survey as quickly as possible, and informed that their
input was important to MSU in serving the agricultural industry in Michigan. To
encourage respondents to mail back the questionnaire, the instrument was designed so that
the respondent could fold it in half, tape it shut, and then mail it without a postage stamp.
A first class mail permit was purchased by the investigator to cover all return postage
costs. The words “BUSINESS REPLY MAIL: Postage will be paid by addressee” were
stamped on the folded “fi'ont” side of the questionnaire. In the upper left hand corner, the

. phrase “No postage necessary if mailed in the United States” was imprinted.

A third cover letter and questionnaire was not used 1) due to the cost of printing and
postage, and 2) because the Michigan agricultural community is a culture that generally
responds quickly or not at all. Less than positive comments written on some
questionnaires by late respondents after the second mailing indicated that a third mailing of
the questionnaire would not be productive. Because a large and adequate number of
questionnaires had been returned at that time, a decision was made not to distribute a third

cover letter and questionnaire.

80

Correcting for Non-response Error
To correct for non-response error, early respondent responses were compared with those
of late respondents (Dillman, 1978; Miller and Smith, 1983; Elliot, 1990). The cut-ofi‘
date used in determining whether a response was “early” or “late” was April 2, 1997. The
final reminder letter and questionnaire were mailed on the afternoon of April 1. Mail in
the United States typically requires one or more days to reach it’s destination. Therefore,
the earliest date that a respondent could have received the second cover letter and
questionnaire in the mail was April 2. Given that a respondent could have immediately
completed the survey, and mailed it back, the earliest postmark that could have appeared
on a returned survey prompted by the final mailing would be April 3, 1997. Therefore,
any postmark on or after April 3, 1997, was considered a “late respondent” for purposes

of this study.

Several additional steps were taken to encourage response and lessen non-response error.
Miller and Smith (1983), state that “The first control strategy involves getting back as
many responses as possible.” Based on Miller and Smith’s recommendations as well as
those by Dillman (1978) and Elliot (1990), a postcard reminder was sent as well as a
second mailing of the questionnaire. A business reply permit was used so that respondents
did not need to purchase a postage stamp. Because firmers do not live close to town or
the local post ofice, the business reply permit allowed the firmer to complete the
questionnaire and then just place it in the mailbox in front of his home or firm. The US.

Mail service allows postal carriers to retrieve out-going mail fiom residential mailboxes

81

and transport it back to the local post office for mailing purposes.

Also according to suggestions by Miller and Smith, oficial Extension letterhead was used
on the cover letters to increase credibility. The postcard and envelop containing the cover
letter and questionnaire included an MSU return address. Also, an MSU address was used
as the location where the questionnaire would be returned, i.e., on the front of the folded
questionnaire. Each cover letter included an original, personal signature in blue ink to add
credibility to the survey process and show personal attention. The cover letter was
examined using WordPerfect Grarnmatik software to make sure that the text was between
sixth and tenth grade reading level, as recommended for general audiences. The
questiomraire was mailed during a less-busy firming season. The questionnaire cover was
printed on colored paper and included a picture of a firm on the front cover. The words
“Michigan Producer” were used on the questionnaire cover instead of “Michigan Farmer”,
as this is the more acceptable and professional term used within the farming industry. All
attempts were made to keep the cover letter and questionnaire short, and the questionnaire

was reduced to 79 percent oforiginal 8.5 by 11 inch paper size.

Data Analysis
Data were compiled and analyzed using SPSS PC+ software. Each survey was examined
by the investigator for valid and legible answers prior to input into the statistical software
package. Any questionnaire that was 1) marked "yes” for actively involved in farming, 2)

included sociodemographic data and 3) had at least one Likert scale item marked were

82

considered useable questionnaires. Others that did not include this information were

considered unusable, with the investigator of the study making this assessment.

All computer input from questionnaires into SPSS was also performed by the investigator
of the study. Each questionnaire had been pre-printed with an identification number which
coincided with one person on the sample mailing list. This information was kept by MASS
and not provided to the investigator to insure confidentiality. As each questionnaire was
received, the identification number was entered into SPSS as an identification number
under a variable labeled “ID”. The identification numbers of those questionnaires received
were then given to the coordinator at MASS, and those firmers were not contacted again.
In addition to the imprinted questionnaire number, each questionnaire was also given a
case number (in pencil) by the investigator in the same order as the responses were loaded
into SPSS. For example, the first questionnaire received and loaded into SPSS was
assigned the case number 001; the second questionnaire was assigned case number 002,
and so on. The questionnaires were then stored for later access in the order of case
number. Having an SPSS “ID” number and corresponding case number provided a

method for quickly accessing stored questionnaires.

A total of 1207 questionnaires were initially mailed. Questionnaires returned due to lack
of a valid mailing address were returned to MASS to re-check for errors or alternate
addresses and re-mailed if possible. After all attempts were made to locate members of

the sample population, it was decided that 94 members of the sample were unreachable,

83

either because they were no longer farming or deceased, or because no forwarding address
was available. Of the 1113 potential respondents remaining, 466 useable surveys were
returned resulting in a 41.9 percent response rate. These questionnaires were entered into
the statistical program and analyzed. Exploratory fictor analytic procedures were utilized,
with a principle components analysis process employed to extract the initial factors
(Guertin and Bailey, 1970; Norusis, 1992; Dunteman, 1989; Harman, 1960; Babbie,
1983). This consisted of an Orthogonal Rotation using the Varimax procedure, an
analysis that reduced the 39 deterrent items in the questionnaire to a smaller number of
“factors”. These “factors” explain most of the variance in the larger set of 39 deterrent
items. The number of fictors to retain was determined using the Kaiser criterion. Factors
with Eigenvalues exceeding 1.0 and having a loading weight of .40 or greater were
maintained for comparisons and correlations, which was similar or identical to criteria

used in other DPS analyses. Those factors retained were use for further statistical

analysis.

Two open-ended questions were provided at the end of the questionnaire. Responses to
the first of these two questions were included in the appendix and used for non-statistical
comparisons with the data. The second open-ended question was provided so that
respondents could provide thoughts or suggestions to the investigator regarding the
questionnaire or the study in general. These comments did not influence the findings in

the study and therefore were not included in the appendix.

CHAPTER 4

REPORT AND ANALYSIS OF THE DATA
Introduction
Chapter 4 presents the findings of this study along with a brief discussion. Included in this
chapter is a demographic summary of the respondents including general deterrent

descriptives, followed by the analytic results to each of the four research questions.

Demographic Summary of the Respondents
Response Rate
A total of 1207 questionnaires were initially mailed. Of these, 94 questionnaires were
returned due to the farmer’s relocation and lack of a new mailing address, retirement from
farnring, or the individual was deceased and no one was able to complete the survey at
that firm. Based on the revised accessible sample population of 1113 firmers, the 466
returned questionnaires resulted in a 41.9 percent response rate. Table 3 shows a
breakdown of respondents by sociodemographic group, including the number of
respondents in each group and percent of total response. The agronomic culture is a
male-dominated society in terms of firm ownership and on-firm jobs, although women

often assume responsibilities as milking or livestock care, equipment operation, financial

84

85

record-keeping for the farm, and other various tasks during peak farming periods. Farm
fimily members tend to carry a strong work ethic, and possess a slightly lower
educational level than a more urban type population (Galbraith, 1992). Therefore it is
logical that sociodemographic response to this survey is more comparable to the US
Census of Agriculture for Michigan than to other populations examined in previous DPS

studies.

Response by Gender

Of those responding to the questionnaire, 98.7 percent indicated their gender. Of these,
400 (87%) were male and 60 (13%) were female. Six respondents did not indicate their
gender. This response rate is somewhat higher for females than the 1992 Census of
Agriculture for Michigan, which shows that of 46,562 farm owners, only 7 percent were
female. However, indication of gender in this study does not represent primary ownership
of the firm, since identified participants in the study were encouraged to pass the
questionnaire on to another individual who was “actively involved in firming” if not able
or qualified to complete it. A male farmer who was quite busy or did not like surveys may
have passed the question on to a female family member who was also actively involved in
the firming business. The survey was not designed to be representative of primary

operator gender, but to describe the educational habits of persons actively involved in
farming.

86

Table 3. Respondents by Sociodemographic Group, Number and

 

 

Percent of Total
Characteristics N“ Valid"
Percent

Gender

Male 400 87.0

Female 60 13.0
Age

18-24 2 0.4

25-34 30 6.5

35-44 87 18.8

45-54 132 28.6

55-64 121 26.2

65 & over 90 19.5
Level of Education (Highest Completed)

Elementary 12 2.6

Some High School 31 6.7

High School Diploma or GED 140 30.4

Some College 86 18.7

Technical or Trade School Certification 50 10.9

2-Year College Degree 38 8.3

4-Year College Degree 36 7.8

Some College Graduate Work 25 5 .4

Graduate Degree 42 9.1

87

Table 3. Respondents by Sociodemographic Group, Number and
Percent of Total (Continued)

 

Characteristics N“ Valid"
Percent
Primary Farming Enterprise
Cash Crops 165 36.3
Dairy 59 13.0
Livestock (except Equine) 88 19.3
Equine 28 6.2
Fruit 34 7.5
Vegetables 18 4.0
Nursery, Greenhouse, Etc. 32 7.0
Other 31 6.8
Annual Gross Sales

Less Than $2500 54 13.7
$2500 - $4999 42 10.6
$5000 - $9999 30 7.6
$10,000 - $24,999 53 13.4
$25,000 - $49,999 50 12.7
$50,000 - $99,999 54 13.7
$100,000 - $249,999 67 17.0
$250,000 - $499,999 21 5.3

$500,000 or more 24 6.1

88

Table 3. Respondents by Sociodemographic Group, Number and

Percent of Total (Continued)

 

Characteristics N* Valid“
Percent

Part-Time or Full-Time Producer

Part-Time Producer 241 52.4

Full-Time Producer 219 47 .6
0&‘-Farm Job

No 239 52.4

Yes 217 47.6
Number of Educational Activities Attended

None 86 24.1

1-5 Activities 191 53.5

6-10 Activities 59 16.5

11 or more Activities 21 5.9

 

 

* Total number of respondents (N) may vary from group to group, due to missing data.

*“ Totals may not equal exactly 100 percent due to rounding error.

89

A T-test for Independent Samples (equality of means with equal variances assumed) was
used to perform the non-response error check for the mail survey. The non-response error
check showed no significant difl‘erence in the early and late respondents. The T-test for
Independent Samples did result in a 2-tailed level of significance exactly at .05 when
applied to the gender variable. Therefore, while gender will be used in this study as an
indicator to summarize toward the general population of females involved in farming in

Michigan, it is possible for an error to exist when using this variable.

Response by Age Category

Age categories were selected based on the 1992 Census of Agriculture. Of those
responding to the questionnaire, 462 (99.1%) indicated the age category in which they fit.
The breakdown included 2 (4%) of the respondents in the 18-24 years category, 30
(6.5%) were ages 25-34, 87 (18.8) were ages 35-44, 132 (28.6%) were ages 45-54, 121
(26.2%) were ages 55-64, and 90 (19.5) were age 65 or over. This is very similar to the
percentage breakdown shown in the Census of Agriculture for Michigan, i.e., 1.3 percent,
9.9 percent, 21.9 percent, 24.2 percent, 21.7 percent, and 20.9 percent respectively for the
same age groups. These results strengthen'the ability to generalize the results of this study

to the general population of Michigan farmers.

Response by Farm Size in Acreage
Farm size ranged from 0.0 acres to 6000 acres. Only one individual indicated he was

actively farming on 0.0 acreage. This person was involved in bee-keeping and it is

9O

assumed that the hives were either placed in another person’s orchards, clover fields, etc.,

or that the hives were located on his/her own property which was less than 1 acre.

A total of 443 (95.1%) individuals responded to this question. Some of those did not
indicate the acreage, while others chose to complete the square footage (greenhouse) item
instead. Some respondents indicated both acreage and square footage. According to the
analysis, the average farm size in acres in the study was 290.04 acres with a standard error
of the mean of 21 .41 acres. This value is higher than the 1992 Census of Agriculture
(1992) which indicates the average farm size in Michigan as 217 acres. However, this
figure is not widely different fiom the 1992 state average, and specialists at the Michigan
Agricultural Statistics Service have stated that their list may omit smaller farms where the
operators may not have identified themselves as “agricultural producers” or “farmers”
(Pscodna, 1996). Also, the 1997 Census of Agriculture is underway and it is expected
that average farm size will increase from that reported in the 1992 Census. Therefore, it
appears that farm size may provide a reasonable base for generalizing data, when used
with a co-variable. Number of acres (farm size) carries little meaning for this study
without relating it to enterprise type. Therefore, this variable was not used in further

analysis for this study, but was retained for any future analyses of the data.

Response by Farm Size in Square Feet
This item was provided for those producers operating greenhouses or nurseries where

facilities or land are measured in square footage, rather than acreage. Of the 466 surveys

91

returned, 32 (6.8 %) respondents indicated farm size by square footage, and size of
operation ranged from 336 to 300,000 square feet. The 1992 Census of Agriculture
shows that in Michigan, about 4 percent of farms are involved in greenhouse and nursery.
The sample response as a percentage exceeded the Census values, however due to the
small number of respondents combined with a wide range in farm size based on square

footage, this sociodemographic criterion was not used further in statistical analysis.

Response by Level of Education

Of the 466 respondents to the mail survey, 460 (98.7%) indicated their highest level of
education completed while 6 respondents did not complete this item. Categories used in
this item of the questionnaire were developed by the investigator to best represent succinct
and easily recognized categories that respondents could identify with. Education level
ranged throughout each category, i.e., from elementary level to those holding a college
graduate degree. Statistical mean for this group was 4.67 or in the category indicating
“some college”, while the mode placed most respondents in the category of “high school
diploma or GE ”. Of the respondents, 12 (2.6%) indicated they had completed
elementary school, 31 (6.7%) had completed “some high school”, 140 (30.4%) held a high
school diploma or GED, 86 (18.7%) had completed “some college”, 50 (10.9%) had
completed technical or trade school certification, 38 (8.3%) had completed a 2-year
college degree, 36 (7.8%) had completed a 4-year degree, 25 (5.4%) had completed some

college graduate work, and 42 (9.1%) had completed a graduate degree.

92

It is interesting to note that 39.8 percent of farmers responding to the survey held the
equivalent of a high school diploma or less, while 60.2 percent had continued their
education beyond high school. It is possible that those with higher levels of education
were more prone to respond to a survey dealing with an educational theme. Regardless,
the number of respondents who had attended trade school or college reflects that farmers
are not an uneducated group. Hobbs (1992) states that “rural communities lag behind in
the proportion of college graduates and in occupations requiring higher levels of education
and training.” He also states that those having higher educational credentials often leave
rural communities to seek jobs which ofi‘er “a good return on their education.” Therefore,
it is expected that educational levels of the farm audience may appear slightly lower than

their urban and suburban counterparts.

Response by Primary Enterprise Type

Respondents represented all categories ofl'ered in the questionnaire with 455 of 466
responding (97.6%). Categories were determined afier reviewing those listed in the 1992
Census of Agriculture and after visiting with Michigan Agricultural Statistics Stafi‘. The
categories selected appeared to best reflect the major agricultural industries in Michigan.
The proportional stratified random sample was based on percentages of farm type in the
MASS mailing list. Respondents to the “Other” category included producers of honey,
wood lots, Christmas trees, and all other agricultural commodities which did not fall into
the other seven enterprise groups. Farmers who were enrolled the USDA Conservation

Reserve Program (CRP) were also included in this category. Although one could argue

93

that CRP is not “actively” farming, the owner is receiving income from farmland for
intentionally keeping it out of production. Therefore, a farmer earning CRP income was

included as a valid respondent who was “actively” farming, and when determining farming

enterprise type.

Of those who responded to the primary enterprise item, 165 (36.3%) produced cash
crops, 59 (13.0%) were dairy producers, 88 (19.3%) raised livestock not including equine,
28 (6.2%) raised equine, 34 (7.5%) were fiuit producers, 18 (4.0%) produced vegetables,
32 (7.0%) listed greenhouse and nursery, etc., and the remaining 31 (6.8%) respondents
selected “other”, as described in the preceding paragraph. The number of questionnaires
mailed and received are listed in Table 4. Response rates by percentage were similar in
proportion to the demographic breakdown in the Census of Agriculture which helps to

support the generalization of the results to Michigan farmers.

Response by Annual Gross Farm Sales

Interval amounts for the “farm sales” item were taken from the 1992 Census of
Agriculture. Of the 466 respondents to the questionnaire, 395 (84.8%) responses were
received, even though this item was listed as “optional”. It was expected that some
respondents might not like to reveal their farm income. Thus in order to encourage
completion of the questionnaire, this item was indicated as an “optic ” item. Each of
the farm sales amounts in this item drew responses. Of the 395 individuals responding, 54

(13.7%) indicated that their average annual farm sales was less than $2500 per year, 42

94

Table 4. Number and Percent of Questionnaires Mailed and
Received by Enterprise Type"

 

 

Number of Number of

Enterprise Type Questionnaires Percent Questionnaires Percent

Mailed Mailed Received Received
Cash Crops 601 49.8 165 36.3
Dairy 148 12.3 59 13.0
Livestock (Exc. Equine) 135 11.2 88 19.3
Equine 61 5.1 28 6.2
Fruit 73 6.0 34 7.5
Vegetables 42 3.5 18 4.0
Nursery, Greenhouse, etc. 76 6.3 32 7.0
Other 71 5.9 31 6.8
Total 1207 100 455" 100

 

"‘ Percentage totals may not exactly equal 100 percent due to rounding error.
‘”" Of the 466 respondents, 455 indicated their primary enterprise.

(10.6%) had farm sales of $2500-$4999, 30 (7.6%) had farm sales of $5000-$9999, 53
(13.4%) had farm sales of $10,000-$24,999, 50 (12.7%) had farm sales of $25,000-
49,999, 54 (13.7%) had farm sales of $50,000-99,999, 67 (17%) had farm sales of
$100,000-249,999, 21 (5.3%) had farm sales of $250,000-499,999, and 24 (6.1%) had

farm sales of $500,000 or more per year.

Response Based on Farming Status
Of the 466 respondents to the questionnaire, 460 (98.7%) respondents completed this

item. Part-time producers out-numbered full-time producers slightly in this study. Of

95
those responding, 241 (52.4%) indicated that they were “part-time producers”, while the
remaining 219 (47.6%) stated that they were farming “full-time”. It should be noted that
part-time firming status does not necessarily reflect that the farmer held an off-firm job.
Some respondents were semi-retired but still operating a farm. Some respondents
indicated that they firm fill-time, while still holding an ofl‘-firm job. It is interesting to
note that the percentages of part-time and full-time firmers appear to be similar to the

percentages in the next item regarding firmers having or not having an off-firm job.

Response Based on OE-Farm Employment

This item collected demographic information related to producer respondents holding, or
not holding, an ofl‘-firm job. Of the 466 total respondents to the questionnaire, 456
(97.9%) responded to this item. Those producers not employed ofl‘ of the firm slightly
out-numbered those with ofl‘-farm jobs. Of those who responded to this item, 239
(52.4%) indicated that they did not have an ofl‘-farm job, while 217 (47.6%) stated that
they did have an ofilfirm job. There were a few questions that arose under this item
related to seed sales from the farm, and whether this was considered ofllfirm employment.
Some producers, especially those with larger firms serve as “seed dealers”, i.e., they serve
as a distribution point for agrononric seed companies. These producers typically keep the
income fi'orn sales or cormnission separately fi'om the firm business. However, since the
income was still derived from the firm facilities, i.e., firm buildings and vehicles used to
store and transport crop seed, etc., an additional job such as a seed dealer was not

considered as an ofl‘-firm job.

96

Participation Behavior
The first item in the questionnaire asked the respondent how many educational activities
he or she had attended in the past 12 months. Of the 466 respondents, 357 completed this
question while 109 did not, i.e., a 76.6 response rate. The lower response to this item may
have been due to the questionnaire structure, which did not number the first question and
thus it may have been overlooked by respondents. However, the validity tests did not
indicate this as a potential problem. More likely, the lower response to this item was due
to the social context of the question, i.e., that the US. culture places a high value on
education and the desire to learn. Individuals who did not attend any educational activities
may have felt it less embarrassing to skip this question than to appear as societal non-

conformists (Krueger, 1994).

Of the 357 individuals responding to this question, 86 (24.1 percent) indicated that they
had not attended an educational program during the past twelve months. Of the remaining
respondents, 191 (53.5 percent) had attended 1 to 5 educational activities, 59 (16.5
percent) had attended 6 to 10 activities, and 21 (5.9 percent) had attended 11 or more
educational activities. In other words, approximately one-fourth of the respondents to this
question had not attended an educational activity during the past year, while three-fourths
had participated, i.e., a three-to-one ratio. The item mean, median, and mode scores for
the respondents were 2.04, 2.00 and 2 respectively, thereby reflecting the category of “1
to 5 activities” attended. This is somewhat similar to the findings of Tough (1971), who

stated that the average person participates in eight learning projects and that in one study,

97

98 percent of the audience surveyed indicated that they had participated in at least one
learning project during the past year. It is interesting to note that this participation and
non-participation rate found in the DPS-F was almost identical to that found by Scanlan
(1982) in the original study using DPS. That study showed that 24.4 percent of the
respondents had not participated in an educational activity in the past 12 months preceding
the study, versus 24.1 percent found in the DPS-F study. Tough's (1971) studies have
shown that the amount of time spent in learning projects varies with type of job, social

class, gender, age, and educational level.

Objective #1: Development of The DPS-F Instrument
Objective # 1 called for the development of a reliable and valid DPS instrument that could
identify firmers’ deterrents to participation in educational programs, i.e., the Deterrent to
Participation Scale for Farmers (DPS-F). Afler face validity and reliability of the pilot
instrument appeared satisfactory, the questionnaire was mailed to farmers and the results
were analyzed. The 39 Likert style deterrent items in the questionnaire were analyzed and
ranked according to mean score and standard deviation (Table 5). Item means range from
1.32 to 3.5, while the overall item mean score was 2.29. The over-all scale mean for the
DPS-F was 89.3. The inter-item correlation mean was .199. Reliability remained high for
the 39 items resulting in alpha = .90. These scores are similar to those found in the

original ups and DPS-G studies.

Of the first ten ranked items, all can be identified as institutional or situational barriers to

98

Table 5. Deterrent Items by Rank, Mean Scores and Standard Deviation
No.

Because the activity was offered during a busy
time of year.

 

2 9. Because the activity was ofl‘ered at an 3.35 1.38
inconvenient location. . ..

3 13. Because ofthe amount oftime required to attend 3.14 1.32
all ofthe sessions...

 

 

 

 

 

 

 

4 14. Because the activity was scheduled at an 3.10 1.42
inconvenient time of day....

5 4. Because the firm or business comes first and I 3.02 1.41
didn’t think I should be gone...

6 23. Because the available activity did not seem useful 2.86 1.44
or practical...

7 3. Because there wasno oneatthefarrnor business 2.86 1.54
who could cover for me...

I 8 8. Because the activity that was ofl‘ered did not seem 2.85 1.31

interesting enough. . ..

9 22. Because I didn’t receive enough information 2.80 1.34
about the actrvr' 'ty ahead of time...

 

 

10 7. Because pre-registration was required and I didn’t 2.75 1.33
want to make a commitment that fir in advance...

11 30. Because I didn’t think the activity would meet my 2.66 1.39

 

 

 

needs...

12 2. Because the activity schedule conflicted with my 2.63 1.65
off-firm job....

13 39. Because I didn’t find out about the activity fir 2.63 1.45

enough in advance to make plans...

 

14 34. Because the activity was not ofl‘ered when I 2.57 1.38
needed the information...

 

 

 

 

 

 
 

Because I didn’t know about educational
activities available on this topic. . ..

Table 5. Deterrent Items by Rank, Mean Scores and Standard Deviation i
Continued
No. i

 
       

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

16 25. Because a last-minute conflict arose that 2.45 1.42
prevented me fl'om attending. . ..

17 32. Because I did not feel that the activity was worth 2.45 1.38
the fee for attending...

18 6. Because I wanted to learn something specific but 2.42 1.26
the activity was too general...

19 21. Because my participation would take away time 2.35 1.31
fi'om my family...

20 5. Because there were too many activities were 2.32 1.32
ofl’ered at the time to attend all of them...

21 28. Because the activity was not on the right level for 2.24 1.30
me. . ..

22 16. Because I prefer to learn through other methods 2.16 1.25
such as reading, radio, etc....

23 17. Because I thought that the activity available 2.14 1.29
would be of poor quality...

24 10. Because I couldn’t afl‘ord the registration or 2.13 1.34
activity fees...

25 24. Because I wasn’t willing to give up my leisure 2.03 1.24
time. . ..

26 27. Because I could afi‘ord miscellaneous expenses 2.01 1.28
like travel, meals, etc...

27 37. Because I was not willing to pay the cost of the 1.98 1.27
registration. . ..

28 19. Because of conflicts with my children’s 1.98 1.34

I activities...
I 29 36. Because I prefer to learn on my own... 1.94 1.22

 

Table 5. Deterrent Items by Rank, Mean Scores and Standard Deviation

100

 

 

 

 

 

 

 

 

 

 

 

 

 

Continued)
_
Rank Item Deterrent Item Mean SB.
No.

30 29. Because the activity conflicted with my civic or 1.89 2.31
community duties...

31 20. Because I’m not that interested in attending 1.84 1.17
educational activities...

32 26. Because I felt unprepared for the activity. . .. 1.77 1.06

33 33. Because I lacked the energy to attend... 1.76 1.11

34 11. Because I felt I was too old to participate in the 1.76 1.20
activity...

35 35. Because no oneIknew was planning to attend... 1.76 1.10 I

36 18. Because I was not confident ofmy learning 1.64 1.10 I
ability...

37 15. Because my family did not encourage my 1.60 .98
participation. . ..

38 31. Because I do not read or write well enough to feel 1.39 .93
comfortable participating. . ..

39 38. Because my fiiends did not encourage my 1.32 .72

artici ation.

 

 

   

 

 

 

101

participation which supports the typology presented by Cross (1981) and assists in

confirming the validity of the instrument.

Prior to performing the fictor analysis (Orthogonal Rotation), Bartlett’s Test of Sphericity
and the Kaiser-Meyer Olkin Measure of Sampling Adequacy were computed to see if the
results were suitable for factorial analysis. Bartlett’s Test gave a chi square of 5940.31
with significance = .00000, while the Kaiser-Meyer Olkin Measure gave a score of .869.
According to Norusis (1992), a factorial model is appropriate when high values occur in
the Kaiser-Meyer Olkin Measure (near 1.0), and when high values in the Bartlett’s Test
correspond with low significance values. Therefore, it was appropriate to perform an
Orthogonal Rotation on the 39 deterrent items in the questiormaire to identify principal

components (Dunteman 1989; Babbie, 1983; Guertin and Bailey, 1970; Harman, 1960).

Using the Varimax Procedure with Eigenvalues accepted at 1.0 or greater, 10 solutions
were originally extracted with loading weights fixed at .40 or higher to define the
solutions. The 10-fictor solution explained 62 percent of the variance. However, to
arrive at a more parsimonious solution, rotations were performed to identify 5, 6, 7, 8, and
9-fictor solutions. After intensive comparison and scrutiny, the 6-factor solution
appeared to provide the most meaningful representation of the data for firrther analysis.
The six fictors accounted for 50. 523 percent of the explained variance. In this solution, 4
of the 39 items loaded on more than one factor. One item which loaded exactly at the .40

level and was retained in order to benefit reliability within that factor group. Two items

102

that did not load under any fictor were omitted in the final solution. After careful
inspection, labels were assigned to each fictor while considering previous fictor labels
described in other DPS studies by Scanlan and Darkenwald (1984), Darkenwald and
Valentine (1985), Davis (1988), Weischadle (1988), and Martindale and Drake (1989).
The DPS-F fictors in order of percent of explained variance were Lack of Course
Relevance, Lack of Confidence, Time Constraints, Low Personal Priority, Scheduling

Conflicts, and Cost (Tables 6-11).

The first fictor, Lack of Course Relevance (Table 6), included eight items related to the
educational activity lacking personal value to the respondent either due to program
content or that it was not ofl'ered at a time when the firmer needed it. This factor label
had been also used by Darkenwald and Valentine (1985) and Martindale and Drake
(1989), and included similar scale items in their fictor sets. In fact, six of the eight items
in the DPS-F study, were identical to those found in both the Darkenwald and Valentine
(1985) study, and the Martindale and Drake (1989) study. In addition, the first and
second items respectively, in the DPS-F findings, “Because I didn’t think the [activity]
would meet my needs” and “Because the available [activity] did not seem useful or
practical”, were also ranked one and two in the Martindale and Drake (1989) study. One
item, “Because I did not feel the activity was worth the fee for atten ' g”, loaded on both
Factor 1 of the DPS-F and on the sixth DPS-F fictor labeled Cost. Loading values
dictated that this item be associated with Factor 1, i.e, Lack of Course Relevance. This

item appeared to have more to do with firrners not valuing the educational activity’s

103

Table 6. Variable Loadings, Item Means, and Scale Ranks:
Factor 1 - Lack of Course Relevance

 

Variable Loading Item Scale
(Item) Value Mean Rank

Because I didn’t think the activity would meet my needs. .808 3.50 11

Because the available activity did not seem useful or .804 2.86 6
practical.

Because I thought the activity available would be of .689 2.14 23
poor quality.

Because the activity that was ofi’ered did not seem .660 2.85 8
interesting enough.

Because the activity was not on the right level for me. .608 2.24 21

Because I did not feel that the activity was worth the fee .608 2.45 17
for attending.

Because I wanted to learn something specific but the .548 2.42 18
activity was too general.
Because the activity was not offered when I needed the .468 2.57 14

information.

 

 

104

Table 7. Variable Loadings, Item Means, and Scale Ranks:
Factor 2 - Lack of Confidence

 

Variable Loading Item Scale
(Item) Value Mean Rank
Because I was not confident of my learning ability. .739 1.64 36
Because I do not read or write well enough to feel .657 1.39 38
comfortable participating.
Because I felt unprepared for the activity. .619 1.77 32
Because I felt I was too old to participate in the activity .604 1.76 34
Because my family did not encourage my participation. .585 1.60 37
Because my fiiends did not encourage my participation. .541 1.32 39

Because no one I knew was planning to attend. .533 1.76 35

 

 

105

Table 8. Variable Loadings, Item Means, and Scale Ranks:
Factor 3 - Time Constraints

 

Variable Loading Item Scale
(Item) Value Mean Rank

Because the firm or business comes first and I didn’t .792 3.02 5
think I should be gone.
Because there was no one at the firm or business who . 763 2.86 7
could cover for me.
Because the activity was ofl‘ered during a busy time of . 592 3.50 1
year.
Because the activity was offered at an inconvenient .499 2.91 9
location.
Because of the amount of time required to attend all of .473 3.14 3
the sessions.
Because there were too many activities ofl‘ered at the .459 2.32 20
time to attend all of them.
Because pre-registration was required and I didn’t want .400 2.7 5 10

to make a commitment that fir in advance.

 

 

106

Table 9. Variable Loadings, Item Means, and Scale Ranks:
Factor 4 - Low Personal Priority

 

Variable Loading Item Scale
(Item) Value Mean Rank

Because I prefer to learn on my own. .757 1.94 29
Because I prefer to learn through other methods such as .682 2.16 22
reading, radio, etc..
Because I wasn’t willing to give up my leisure time. .629 2.03 25
Because my participation would take away time fiom my .549 2.35 19
fimily.
Because I’m not that interested in attending educational .487 1.84 31
Because I lacked the energy to attend. .472 1.76 34

 

 

107

Table 10. Variable Loadings, Item Means, and Scale Ranks:
Factor 5 - Scheduling Constraints

Variable Loading Item Scale
(Item) Value Mean Rank

Because the activity schedule conflicted with my ofl’-farm .633 2.63 12
job.

Because I didn’t find out about the activity fir enough in .570 2.63 13
advance to make plans.

Because of conflicts with my children’s activities. .556 1.98 28
Because I didn’t receive enough information about the .547 2.80 9
activity ahead of time.

Because the activity was scheduled at an inconvenient .429 3.10 4
time of day.

Because I didn’t know about educational activities .440 2.51 15

available on this topic.

 

 

108

Table 11. Variable Loadings, Item Means, and Scale Ranks:

 

 

Factor 6 - Cost
Variable Loading Item Scale
(Item) Value Mean Rank
Because I was not willing to pay the cost of the .819 1.98 27
registration.
Because I couldn’t afi’ord the registration or activity fees. .802 2.13 24
Because I couldn’t afi‘ord miscellaneous expenses like .787 2.01 26

travel, meals, etc..

 

 

relevance to their needs (willingness to pay for it), than a problem of the activity not being
afl’ordable (ability to pay for it). A second item, “Because the activity was not offered
when I needed the information” also loaded on Factor 1 and on Factor 5, Scheduling
Constraints. However, the item loaded more heavily under Factor 1, and the aspect of
program relevance seemed to be a more appropriate fit than conflict with farmers’
schedules. Therefore, this item was also included under Lack of Course Relevance.

Factor 1, accounted for 11.7 percent of the total explained variance, with loading weights

ranging fi'om .808 to .468. Item-total reliability was alpha = .8591 for the factor group.

Factor 2 identified by the analysis was termed “Lack of Confidence”, which had also been

identified in previous DPS studies (Darkenwald and Valentine, 1985; and Martindale and

109

Drake, 1989). Seven items, shown in Table 7, were included in this factor group with
loading weights ranging from .739 to .533. Identified deterrents related to the farmer not
feeling prepared, feeling too old to participate, not receiving encouragement from family
and/or fiiends to attend, etc.. The item, “Because I was not confident of my learning
ability” appeared as the most important item in the DPS-F results, and in both the
Darkenwald and Valentine study and the Martindale and Drake study, based on loading
values within the fictor group. Five of the items appearing in the DPS-F study also
appeared in the Darkenwald and Valentine (1985) study. One new item added to the
DPS-F study, i.e., “Because I do not read or write well enough to feel comfortable
participatin ”, emerged as the second highest loading item in this fictor. This item was
developed and added during the DPS-F focus group input sessions with Michigan
producers, and helps confirm the importance of focus group input and the validity of the
instrument. The strong similarities in item content within the fictor group to other DPS
studies led to the assignment of the label, Lack of Confidence. The item “Because I’m not
that interested in attending educational activities” loaded on both Factor 2, Lack of
Confidence and Factor 4, Low Personal Priority. After reviewing respective loading
weights and the general meaning of the deterrent item, it was retained under Factor 4,
Low Personal Priority. Factor 2 accounted for 9.102 percent of the total explained

variance, with an item-total reliability for the group of alpha = .7834.

Factor 3 was labeled Time Constraints (Table 8). This fictor label is identical to that used

in four previous DPS studies (Darkenwald and Valentine, 1985; Davis, 1988; Weischadle,

110

1988; and Martindale and Drake, 1989) and relates to the firrners’ inability to attend due
to time limitations. Seven items loaded under this component with loading weights
ranging from .792 to .400. The highest loading item, “Because the farm or business
comes first and I didn’t think I should be gone”, was a new item added to the DPS-F
based on focus group input, again confirming the importance of focus group input and the
validity of the instrument. Other items dealt with the time required to attend all of the
sessions, the activity being held during a busy time of year or inconvenient time of day,
etc.. Another new item added to the DPS-F, “Because there were too many activities
offered at the time to attend all of them”, emerged in this factor set. The item, “Because
pre-registration was required and I didn’t want to make a commitment that fir in advance”
loaded exactly at the .40 level in this factor group, and was retained to benefit the alpha
score. One item, “Because the activity was scheduled at an inconvenient time of day”
loaded on both Factor 3, Time Constraints, and under Factor 5, Scheduling Constraints.
Thisiternappearedtobeabetterfitunder SchedulingConstraintssinceitdealtmorewith
conflicts of schedules than the firmer having the time to participate. Alter comparing
item-total reliability using both scenarios, a decision was made to place this item under
Factor 5, Scheduling Constraints. The item, “Because a last-minute conflict arose that
prevented me fiom attending” appeared under Factor 3, Time Constraints however, this
item scored below the required loading weight of .40 and therefore was omitted fi'om the
results. Item-total reliability for Factor 3 was alpha = .7613, with a total explained

variance of 8.598 percent for the fictor group.

111

Factor 4 was labeled “Low Personal Priority”. This factor group, shown in Table 9,
included six deterrent items with loading weights ranging from .757 to .472. This
principal component included items such as “Because I prefer to learn on my own”,
“Because I lacked the energy to attend”, “Because I wasn’t willing to give up my leisure
time”, etc.. This factor group was most similar to Darkenwald and Valentine’s fictor
group titled “Low Personal Priority”, but contained two items respectively in both
Martindale and Drake’s factor group labeled “Lack of Interest” and Scanlan and
Darkenwald’s factor group labeled “Disengagement”. Because there was a closer match
with Darkenwald and Valentine’s fictor group, and because all of the items loading under
this fictor can be attributed to a decision based on “low personal priority”, that label was
assigned to the component. The item “Because I’m not that interested in attending
educational activities loaded on both Factor 2, Lack of Confidence and Factor 4, Low
Personal Priority. After reviewing respective loading weights and the general meaning of
the deterrent item, it was retained under Factor 4. Factor 4 accounted for 7.587 percent

of the total explained variance, with a item-total reliability of alpha = .7494.

Factor 5 dealt with the issue of conflicts between the firmers schedule or commitments
and the inability to participate. Five deterrent items loaded under this fictor with weights
ranging fiom .633 to .440 (Table 10). Items such as conflict with an off-firm job or
children’s activities, and not finding out about the program in time, provided the focus for
this component. As a result Factor 5 was labeled “Scheduling Constraints”. There

appears to be a “situational” or “institutio ” theme as the foundation for this fictor

112

group. Inability to attend is not controlled by the farmer but is based on his situation at
the time the program was ofl‘ered or announced. Weischadle (1988) also used the label
Scheduling Constraints in her results, although the items loading under the DPS-F group
appeared to be an even better fit for the label than some of those found in her study. Of
the five DPS-F items loading, four of them are identical or directly related to items in
Weischadle’s findings. The fifth item, “Because of conflicts with my children’s activities”
was not inchrded in Weischadle’s findings but is clearly an issue of schedule conflicts and
fits well in this group. As stated under Factor 3 above, one item, “Because the activity
was scheduled at an inconvenient time of day” loaded on both Factor 3, Time Constraints
and under Factor 5, Scheduling Constraints. A comparison of item-total reliability using
both scenarios, as well as “apparent fit”, showed that this item should be placed under
Factor 5. One item, “Because the activity conflicted with my civic or community duties”
was extracted under Factor 5, however it did not attain the required .40 loading criterion
and was omitted from the results. Explained variance for Factor 5 was 6.812 percent of

the total explained variance, while item-total reliability for the group was alpha = .6904.

The final fictor set, Factor 6, was labeled Cost. Three items loaded under this component
with each item referring to the ability or desire to pay for the activity. Loading weights of
the deterrent items in this component ranged fiom .819 to .787 (Table 11). These
included “Because I was not willing to pay the cost of registration”, “Because I couldn’t
afl’ord the registration or activity fees”, and “Because I couldn’t afford miscellaneous

expenses like travel, meals, etc.” This “Cost” fictor group emerged in all five previous

113

DPS studies surficing either in the third, forth, or fifth order of importance based on
amount of explained variance. Cross (1981) stated that having the ability to pay is not the
same as having the willingness to pay. The second and third items loading in this
component clearly reflect an issue of “ability to pay”. The first and heaviest loading item
“Because I was not willing to pay the cost of the registration” could be seen as a
determination that the activity was not valued enough for it’s relevance to warrant paying
the registration; or it might instead be a statement by the farmer that he could not afi‘ord to
pay for educational meetings as long as the information is available elsewhere at less cost.
This item did not load on Factor 1, Lack of Course Relevance, and therefore it is assumed
that “cost” of the activity in this factor is more related to inability to pay and therefore was
a satisfactory fit for the Cost fictor. Another cost-related item, “Because I did not feel
that the activity was worth the fee for attending” loaded both under Factor 1, Lack of
Course Relevance and Factor 6, Cost. This item loaded more heavily on Factor 1, and it
clearly reflected that the firmer might have paid a fee if the activity would have been
“worth” the fee being charged. Therefore, that item was included in Factor 1 to match the
best fit for the two components. Explained variance for Factor 6 was 6.710 percent of

total explained variance, while item-total reliability was alpha = 8330.

In summary, the six factors that emerged fiom the Orthogonal Rotation appeared to be
parsimonious, and provide support to the validity of the instrument in its ability to identify
relevant deterrents for firmers. In addition, alpha coeficients of reliability for total item

scores and for fictor item scores were acceptable and similar to other DPS scales,

114

particularly the DPS-G which was used as the primary template. Table 12 lists the

resulting six fictors by label in order of explained variance and cumulative variance.

Table 12. Deterrent Factors in Order of Importance by Explained

Variance and Cumulative Variance

 

 

 

 

 

 

FACTOR FACTOR EXPLAINED CUMULATIVE
NUMBER LABEL VARIANCE VARIANCE

1 Lack ofCourse Relevance 11.715 11.715

2 Lack of Confidence 9.102 20.817

3 Time Constraints 8.598 29.415

4 Low Personal Priority 7.587 37.002

5 Scheduling Constraints 6.812 43.814

6 Cost 6.710 50.523

 

Research Question #1: Factor Difl’erentiation

Research Question #1 calls for a comparison of the six fictors identified by Orthogonal

Rotation, to determine if they are statistically distinct and separate source variables. A

General Linear Model was applied using a Repeated Measures test, with the deterrent

factors serving as the dependent variables and within-subject factors. Mean scores and

standard deviations of the six fictors are listed in Table 13. A multivariate analysis

showed that Pillai’s Trace, erks’ Lambda, Hotelling’s Trace, and Roy’s Largest Root

tests all resulted in a significant difference between the means of the six factors at p <

 

115

.001. Therefore, statistical analysis does demonstrate that a significant difference does
exist between the six fictor mean scores. These results provide statistical evidence that
the six source variables (fictors) identified through Orthogonal Rotation are distinct and

separate, and may be used to further analyze the data with respect to the deterrent fictors.

Table 13. Repeated Measures Test to Compare Factor Scores"

 

 

 

 

 

 

 

 

FACTOR" MEAN S.D. N
3-Time Constraints 2.9996 .8821 446
5-Scheduling Constraints 2.4999 .9241 446
l-Lack of Course Relevance 2.4640 .9527 446
6-Cost 2.0419 1.1239 446
4-Low Personal Priority 2.0097 .8247 446
2- Lack of Confidence 1.6070 .6972 446
"‘ p < .001

* *Factors are listed in order of mean score rather than factor identification number.

Research Question 2: Deterrent Factors And Sociodemographic Variables
Correlations for Factors and Sociodemographics
Research Question 2 asked whether deterrent fictors varied in type and importance with
varying sociodemographic variables. In order to address the question, the six source
factors were first correlated with sociodemographic variables using Pearson Correlation
Coefi'rcient (r) calculations. Although a Spearrnan Rank-Order analysis is typically used to
compare relationships between dichotomous category-type variables (Elliot, 1991 ), a

Pearson analysis can be substituted if resulting values are essentially the same. A sample

116

test between the strongest correlating variables, i.e, between Scheduling Constraints and
Off-firm Job yielded correlations of r = .334 (p < .01) for the Pearson Correlation and r =
.340 (p < .01) for the Spearman non-parametric correlation The two processes resulted
in similar values and therefore, the Pearson Correlation was used for analyzing preliminary
relationships between the fictors and sociodemographic variables. This analysis was used
only as an indicator for firrther ANOVA tests. Primary enterprise (commodity type) was
not used in the correlation due to it’s nature as a multi-categorr'cal variable. Table 14

depicts the resulting correlations.

The sociodemographic variable, gender, correlated negatively with Low Personal Priority
and positively with Cost. A positive relationship was shown between age and Lack of
Confidence, while Time Constraints and Scheduling Constraints resulted in a negative
relationship with age. A negative relationship occurred between level of education and
Low Personal Priority. Annual gross sales correlated positively with Lack of Course
Relevance and Time Constraints, and negatively with Low Personal Priority and Schedule
Constraints. Farnring status (part-time versus fill-time) showed a positive correlation
with Time Constraints and a negative correlation with Low Personal Priority and
Scheduling Constraints. Holding an off-firm job correlated negatively with Lack of

Confidence and positively with Scheduling Constraints.

The strongest correlation occurred between Factor 5, Scheduling Constraints and holding

an off-farm job with a Pearson r-value of .334 (p < .01), i.e, a moderate association

117

Table 14. Correlations Between Source Factor Scores and Sociodemographic

 

 

Variables
Source Factors
Socio- l 2 3 4 5 6
mmmphic Lackof Lackof Time Low Scheduling Cost
Variables Course Confi- Con- Personal Constraints
Relevance deuce straints Priority
Gender -.022 -.O45 -.O33 -.114* -.O36 .160“
Age .011 .201 ** -.107"‘ .072 -.109"' .003
Level of -.015 -.205“ -.O42 -.115"‘ .035 -.029
Education
Annual .146" -.090 .226" -.122* -.154" -.052
Gross Sales
Part-time or .036 -.040 .188" -.129** -.213"“' .016
Full-time
GIT-Farm -.016 -.145"'"‘ —.002 -.015 .334" -.034
Job
*p < .05 IMIp < .01

118

(Elliot, 1991). Annual gross sales correlated significantly with four factors. In three of

these four correlations, significance appeared at p < .01.

ANOVA’s for Factors And Sociodemographics

One-Way Analysis of Variance was applied to test difl‘erences in the type and importance
of the six deterrent fictors, based on sociodemographic groups where significant Pearson r
correlations appeared. Tukey’s Honestly Significant Difi‘erence (HSD) test provided post
hoc range tests and pairwise multiple comparisons where categories within the
sociodemographic variable numbered more than two. Each deterrent factor served as a
separate dependent variable. Where mean scores indicated a difi‘erence of .5 or greater,

the difi‘erence was stated to be both “significant and meaningful”.

Two of the sociodemographic variables having eight or nine categories, were transformed
by condensing categories into fewer logical groups before the ANOVA was applied.

Level of Education was reduced from nine categories to four, i.e., holding less than a high
school diploma, holding a high school diploma, at least some post-secondary education,
and 4-year college degree or beyond. Annual gross sales was also reduced fiom nine
categories to three, i.e, $9,999 or less per year, $10,000 to $99,999 per year, and
$100,000 or more per year. Number of acres was not used in the comparisons because
the “number of acres” on a firm does not reflect level of income, level of management
required, etc., and thus it is dificult to interpret the outcomes. For example, a firmer with

muck vegetables may be an intensive, firll time farmer with only 30 acres, while a cash

119

crop firmer could not sustain fill-time employment or a livelihood on only 30 acres.

Only two firmers in the youngest category (18-24) responded to the questionnaire. Elliot
(1991) stated that a sample is unbiased when N is equal to or greater than 25, therefore
the 18-24 and 25-34 age categories were combined for further analysis with n = 32. The

remaining four categories were left unchanged.

Results of ANOVA: Lack of Course Relevance

Factor 1, Lack of Course Relevance demonstrated a low but positive association with
annual gross sales at r = .146 (p < .01). The trend appearing in the data suggests that
course relevance becomes more important as annual gross sales increases. This appears to
be logical, since with increasing sales comes increasing time spent in managing the farm
operation. Time would likely be less available and those who would attend educational
activities may put more emphasis on attending only those activities that were most

relevant to their needs. However, analysis of variance (Table 15) showed that while a
pattern exists, there was no significant nor meaningful difl’erence between the three income
groups (p = .066). No other sociodemographic group displayed a significant correlation

with this factor.

120

Table 15. Summary of AN OVA: Lack of Course Relevance and Annual Gross

 

 

Sales"

Annual Gross Sales Mean S.D. N
$9,999 or less 2.3202 .9520 117
$10,000 - $99,999 2.5279 .9516 152
$100,000 or more 2.5964 .9070 110
Total 2.4837 .9433 379

I"ANOVA did not show a significant difference at p < .05

Results of AN OVA: Lack of Confidence
Factor 2, Lack of Confidence demonstrated a significant but low association with age at r
= .201 ( p <.01), level of education with r = -.205 (p < .01), and holding or not holding an

ofl‘-farm job with r = -. 145 (p< .01).

When an analysis of variance was applied to Lack of Confidence and the age categories,
the significance between groups resulted in p < .001 (Table 16). Lack of Confidence was
compared with the means of the six age groups using Tukey’s HSD test. A significant
difl'erence was seen in the importance of confidence as a deterrent between those ages 65
and over with those in the 18-34 year old category (p < .05), and the 35-44 category (p <
.05). A significant difl‘erence was also seen between the 45-54 category and the 55-64 age
group (p < .001). While each was significant, the only meaningful difi’erneces were
between the 18-34 age group and 65 and over age group, and between the 45-54 age

group and 65 and over age group.

121

Table 16. Summary of ANOVA: Lack of Confidence and Age

 

 

 

 

Age Mean S.D. N
18-34 1.5144 .4806 32
3544 1.5567 .5694 87
45-54 1.4740 .6441 132
55-64 1.7486 .9003 117
65 and over 2.0146 1. 0523 85
Total 1.6651 .8067 453
-—m—n
:uares .
Between 17 768 4 442 ‘
Groups
Within 276.408 448 .617
Groups
Total 294.176 452

 

 

 

 

 

An analysis of variance was applied to Lack of Confidence and level of education (Table
17). Examining mean scores, a trend does appear in the data showing a general increase
in Lack of Confidence as a deterrent as the level of education decreases. This would be
expected and compares with findings by Darkenwald and Merriam (1982). However,
while the correlation showed a low association i.e., r = -.205 (p <.001), significant but not
meaningful difl‘erences appeared in the ANOVA between firmers possessing a 4-year
degree or higher with those with no high school diploma, and between farnrs with a 4-year

degree or higher with those holding a high school diploma.

122

Table 17. Summary of ANOVA: Lack of Confidence and Level of Education

 

Level of Education Mean S.D. N
Less than High School Diploma 1.8542 .8097 41
High School Diploma or GED 1.8342 .9118 137
Some College 1.6329 .6678 171
4-Year Degree or more 1.3868 .7346 102
Total 1.6585 .7931 451

 

-_$—n
“W

13.443

 

269.612

 

283.055

 

 

 

 

 

 

Lack of Confidence also demonstrated a negative correlation with holding or not holding
an off-farm job at r = -.145 (p < .01), although only displaying a low association. Table
18 includes the results of the analysis of variance showing a significant difi‘erence in the
two categories within the sociodemographic group (p < .05). However the deviation in

means was not sufficient to show a meaningfirl difference between the groups.

123

Table 18. Summary of ANOVA: Lack of Confidence and Off-Farm Job

 

Off-Farm Job Mean S.D. N
No Off-Farm Job 1.7803 .9140 232
Off-Farm Job 1.5446 .6628 215
Total 1.6670 .8108 447

 

 

-—m—n
uares

 

 

 

 

 

‘ Between 6. 200 6. 200 9. 614
Groups
Within 286.971 445 .645

. Groups
Total 293.171 446

 

 

 

 

 

 

Results of AN OVA: Time Constraints

Factor 3, Time Constraints showed a negative correlation with age of p < .05, and positive

correlations with both annual gross sales and farming status (part-time versus fill-time) at

p < .01. Table 19 portrays the ANOVA between Time Constraints and age. Group mean

scores of those in the five age categories showed that those in the “65 and over” category

indicated Time Constraints to be less important as a deterrent than the other age groups.

This appeared to be especially true when comparing farmers age 65 and over, with farmers

age 35-44 (p = .052). This also compares with studies by Darkenwald and Merriam.

124

However, for farmers, the Tukey HSD test did not show a significant nor meaningful

difference between age groups.

Table 19. Summary of ANOVA: Time Constraints and Age"

 

Age Mean S.D. N
18-34 2.9844 .7076 32
35-44 3.1743 .6998 87
45-54 3.0854 .8414 132
55-64 3.0042 .9296 119
65 and over 2.7857 1.1157 82
Total 3.0196 .8939 452

 

‘ANOVA did not show a significant difference at p < .05

The Time Constraints fictor showed a positive correlation with annual gross sales at r =
.226 (p < .01). The correlation showed a direct relationship between groups with p =
.000. Time Constraints became more important as a deterrent with increasing income. An
ANOVA was used to examine the differences between the three income levels within this
category (Table 20). Tukey’s test indicated a significant difl'erence (p < .001) between
the lower income level ($9,999 or less) and the higher income level ($100,000 or more),
but not between the lower and middle income levels. A significant difl’erence also
appeared between the middle level ($10,000 to $99,999) and higher level with p < .05. A
meaningful difference occurred between the lower gross sales group and higher gross sales

group, but not between the middle and higher groups.

125

Table 20. Summary of ANOVA: Time Constraints and Gross Annual Sales

 

 

Annual Gross Sales Mean S.D. N
$9,999 or less 2.7751 .9270 121
$10,000 - $99,999 3.0055 .8803 153
$100,000 or more 3.2837 .6585 111
Total 3.0133 .8594 385

14.990

 

268.599

 

283 .589

 

 

 

 

 

 

Time Constraints also displayed a low but positive association with firming status, i.e.,
farming part-time or full-time with r = .188 (p < .01). The analysis of variance resulted in
a significance level of p < .001. Mean score comparisons show that lull-time producers
scored Time Constraints as more important as a deterrent to participation than did part-

time farmers (Table 21), however a meaningful difference did not exist.

Results of ANOVA: Low Personal Priority
Factor 4, Low Personal Priority showed a low association but did correlate negatively

with gender, level of education, annual gross sales, and firming status.

126

Table 21. Summary of ANOVA: Time Constraints and Farming Status

 

FarmingStatus Mean S.D. N
Part-Time Producer 2.8651 .8799 235
Full-Time Producer 3.1957 .8518 215
Total 3 .0231 .8813 450

 

12.267

 

336.459

 

348.726

 

 

 

 

 

 

Primary enterprise, although not included in the Pearson correlation, did reveal significant
difl‘erences in the ANOVA test between dairy enterprises and four other enterprise groups
including cash crops, livestock, fruit, and “other” at p < .01. Mean score difi’erences show
that dairy producers report time as a more important deterrent than producers fiom any of
these other four enterprise groups. The difi’erences in mean scores between these groups
also demonstrate that the differences are statistically meaningful. Equine (n = 25) and
vegetables (n=18) did not provide large enough samples to adequately eliminate statistical

error, and therefore were not considered in the ANOVA results.

127

Table 22. Summary of ANOVA: Time Constraints and Primary Enterprise

 

 

 

 

 

 

 

 

 

Enterprise Mean S.D. N
Cash Crops 2.8543 .9215 162
Dairy 3.5764 .7879 58
Livestock 2.9620 .8748 87
Equine 3.0350 .7131 25
Fruit 2.8346 .8949 34
Vegetables 3.2917 .7110 18
Nursery, etc 3.1992 .7987 32
Other 2.8869 .8903 30
Total 3 .0224 .8927 466
-_nm—u
. uares
Between Groups 26.752 3. 822 5.105
Within Groups 327.911 438 .749
Total 354.663 445

 

Low Personal Priority and gender correlated with low association at r = -.114 (p < .05).
An analysis of variance between these two variables showed a significant difi‘erence
between males and females in their scoring (p < .05). Mean score comparisons
demonstrate that males tend to find this fictor as more important than females as a

deterrent to participation (Table 23), although not at a meaningful level.

128

Low Personal Priority and level of education also correlated with low, negative

association at r = -.115 (p < .05). There appears to be some preference for those with a

higher level of education to have more interest in participating and less concern about how

Table 23. Summary of ANOVA: Low Personal Priority and Gender

 

 

Gender Mean S.D. N
Male 2.0826 .8829 392
Female 1 .7819 .8054 57
Total 2.0444 .8783 449

-nm—n
I!“

 

 

V Between 4.500 4.500
Groups
Within 341 . 129 447 .763
. Groups
' Total 345.628 448

 

 

 

 

 

they learn. This would likely be expected. An analysis of variance resulted in a

significant difl°erence between groups of p = .030, while Tukey’s HSD test showed

significance at p = .044 for mean difl’erences between those who had completed some

college and those who had completed a 4-year degree (Table 24). However, based on the

resulting mean scores, no meaningful difi‘erence exists between education levels.

129

Low Personal Priority correlated with annual gross sales, although a low, negative

association appeared at r = -.122 with p < .05. Mean scores show a definite inverse

relationship, however, an analysis of variance demonstrated that while these two variables

Table 24. Summary of AN OVA: Low Personal Priority and Level of Education

 

Level of Education Mean S.D. N
Less than High School Diploma 2.2114 1.0730 41
High School Diploma or GED 2.0694 .9039 135
Some College 2.1125 .8360 170
4-Year Degree or more 1.8284 .7655 102
Total 2.0439 .8716 448

 

nm—n
. uares

 

 

 

 

 

 

Between 6. 774 2. 258 3. 012
Groups

' Within 332.836 444 .750

Groups

I Total 339.610 447

 

are related, categories within the annual gross sales variable did not show a significant

difi‘erence between means (Table 25). Although not large enough to be meaningful, there

does appear to be a slight trend of decreasing importance in the deterrent of Low Personal

Priority as income increases.

130

Factor 4, Low Personal Priority and farming status correlated with low, negative
association at r = -.129 although with higher significance than other sociodemographic

variables (p < .01 ). An analysis of variance procedure revealed a significance in

Table 25. Summary of ANOVA: Low Personal Priority and Gross Annual Sales“

 

Annual Gross Sales Mean S.D. N
$9,999 or less 2.1924 .9494 119
$10,000 - $99,999 2.0249 .8803 155
$100,000 or more 1.9606 .6814 110
Total 2.0584 .8545 384

 

‘ANOVA did not show a significant difference at p < .05

difi’erences of means for those farming part-time and those farming firll-time (p = .006).
Table 26 shows the mean difl‘erences and standard deviations for this relationship.
Although not large enough to be considered meaningful, a slight pattern exists showing
that part-time farmers see Low Personal Priority as more important in deterring them fiom

participating in educational activities than full-time firrners.

Results of ANOVA: Scheduling Constraints

Factor 5, Scheduling Constraints gave the highest number (4) of correlations with the six
sociodemographic variables, relating significantly to age, annual gross sales, firming
status, and ofl’-farm jobs. Correlation Coeficients all showed relationships ranging fiom r

= -.109 (p < .05) to r = .334 (p < .01), the latter being the strongest correlation among the

131

Table 26. Summary of ANOVA: Low Personal Priority and Farming Status

 

Farming Status Mean S.D. N
Part-Time Producer 2.1613 .9512 235
Full-Time Producer 1.9352 .7734 213
Total 2.0538 .8776 448

 

 

5.710

 

338.535

 

344.245

 

 

 

 

 

 

factors and sociodemographic variables.

The correlation of Scheduling Constraints and age revealed a low, negative association of
r = -.109 (p < .05), which was the smallest correlation under this factor. Table 27
provides the results of the AN OVA and Tukey test. The Tukey test revealed that a
significant difl'erence in mean values appeared between firmers in the 65 and over
category and those in the 35-44 age group, and between the 65 and over category with the
45-54 age group. The data imply that farmers in the oldest age group see scheduling as
less of a deterrent to participation than those in mid-career, but no difi‘erence appears

between any other groups. No meaningful difference exists between the age groups.

132

Table 27. Summary of ANOVA: Schedule Conflicts and Age

 

ége Mean S.D. N
18-34 2.3958 .8639 32
35-44 2.6981 .9315 87
45-54 2.6572 .9109 132
55-64 2.5034 .9916 118
65 and over 2.2463 .9567 81
Total 2.5322 .9518 450

 

11.770

 

395.021

 

406. 792

 

 

 

 

 

 

The correlation of Factor 5, Scheduling Constraints and annual gross sales resulted in a
value of r = -.154 (p < .01), a negative and low association between the variables. When
an ANOVA was performed on these variables (Table 28), a significant difl’erence surfaced
between the lower income group (less than $9,999 per year) and the higher income group
($100,000 and over per year). However, mean scores did not difi‘erentiate sufliciently to
demonstrate a meaningful difi‘erence due to gross farm sales. No difi’erence was seen

between any other two groups.

133

Table 28. Summary of ANOVA: Schedule Conflicts and Annual Gross Sales

 

 

 

 

Annual Gross Sales Mean S.D. N
Less than $9,999 2.7292 1.0194 120
$10,000 - $99,999 2.5730 .9444 152
$100,000 or more 2.3214 .8861 112
Total 2.5484 .963 1 384

-—m—n
snares

f Between 9. 783 4. 892 5.394

Groups

Within 345.490 381 .907

Groups

Total 355.273 383

 

 

 

 

 

 

Factor 5, Scheduling Constraints, also correlated significantly with firming status showing
a low, negative association at r = -.213 (p < .01). This relationship was the second
strongest of all comparisons between factors and sociodemographic groups. Analysis of
variance showed that part-time producers identified Scheduling Constraints significantly
more important as a deterrent to participation than full-time firrners (p < .001). Still, with
mean scores differing by only .41, there was not an adequate interval to indicate that a
meaningful difl’erence existed. Table 29 shows a comparison of means and standard

deviations as well as the results of the ANOVA test.

134

Table 29. Summary of ANOVA: Scheduling Conflicts and Farming Status

 

 

Farming Status Mean S.D. N
Part-Time Producer 2.7323 .9547 235
Full-Time Producer 2.3280 .9004 213
Total 2.5401 .9500 448

18.259

 

385.168

 

403.427

 

 

 

 

 

 

The final correlation between the fictor Scheduling Constraints and “OE-farm job”
resulted in the highest correlation between the six factors and the sociodemographic
variables with r = .334 (p < .01). The correlation resulted in a positive relationship and
was the only correlation among the variables resulting in a “moderate” association. The
data fiom the ANOVA (Table 30) show that firmers who hold an ofi-farm job have more
difficulty with Scheduling Constraints as a deterrent to their participation in educational
activities than finners who do not hold an ofl‘-farm job. Results of the analysis of

variance were both significant and meaningful with p < .001.

Table 30. Summary of ANOVA: Scheduling Constraints and Off-Farm Job

135

 

Off-Farm Job Mean S.D.
No Off-Farm Job 2.2196 .9099
Off-Farm Job 2.8509 .8751
Total 2.5282 .9464

 

 

-_—$_—.
III!”

 

 

 

 

Results of ANOVA : Cost

Factor 6, Cost, was the final source variable correlated with sociodemographic variables
and the only sociodemographic variable resulting in a significant correlation with gender at
r = .160 (p < .01). A positive, but low association exists. The ANOVA shows that cost is
significantly more important to females as a deterrent that it is to males, and a meaningful

difl‘erence exists between the group mean scores. Significance resulted in p < .001 (Table

31).

 

 

 

Between 44.218 44.218 55.437
Groups

\Vrthin 352.549 442 .798

Groups

Total 396.767 443

 

136

Table 31. Summary of ANOVA: Cost and Gender

 

 

 

 

Gender Mean S.D. N
Male 1 .9742 1 .0927 387
Female 2.5179 1.2727 56
Total 2.0429 1.1300 443
-———n
. uuares S uare
Between 14.461 14461 11.597
Groups
\Vrthin 549.946 441 l .247
Groups
. Total 564.407 442
{

 

 

 

 

 

Research Question 3: The Importance of Cost as a Deterrent
Research Question #3 asked if cost was the most important deterrent to participation in
educational activities for farmers. To address this question, it was necessary to
statistically rank the six deterrent fictors based on the individual rankings. This ranking
was performed using a non-parametric Friedman’s Test which tests the null hypotheses
that k related variables come from the same population. The test was performed using the

entire sample of firmers who responded to this item (Table 32).

137

Table 32. Friedman’s Test to Rank the Deterrent Factors:
Participants and N on-Participants

 

 

 

 

 

 

 

 

 

 

 

 

MEAN FACTOR
FACTOR RANK RANK ITEM MEAN S.D. N
Time Constraints 1 5.01 2.9996 .8821 446
Scheduling Constraints 2 3.98 2.4999 .9241 446
Lack of Course Relevance 3 3.93 2.4640 .9527 446
Low Personal Priority 4 3.10 2.0097 .8247 446
Cost 5 2.88 2.0419 1.1239 446
Lack of Confidence 6 2.11 1.6070 .6972 446
I N I Chi-Square I df I Sig. I
I 446 I 697.405 I 5 I .000 I

 

Results of the Friedman’s Test showed that the deterrents were statistically different at p <
.001 and ranked them in the following order fiom most important to least important: 1)
Time Constraints, 2) Schedule Conflicts, 3) Lack of Relevance, 4) Low Personal Priority,
5) Cost, and 6) Lack of Confidence. A Friedman’s Test was then performed for non-
participants only (Table 33) with identical results in rank order of deterrents except for
Lack of Relevance and Low Personal Priority. These two fictors interchanged between
third most important and fourth most important. Levels of significance again emerged as
significant (p < .001). However, in either test it is shown that the deterrent factor “Cost”

is not the most important and remains fifth in importance based on Friedman’s ranking.

138

Table 33. Friedman’s Test to Rank the Deterrent Factors: Non-Participants

 

 

 

 

 

 

 

 

 

 

 

 

MEAN FACTOR

FACTOR RANK RANK ITEM MEAN S.D. N
Time Constraints 1 4.66 2.6376 1.0363 78
Scheduling Constraints 2 3.67 2.2692 .9459 78
Low Personal Priority 3 3.60 2.1137 .8939 78
Lack of Course Relevance 4 3.44 2.1548 .8880 78
Cost 5 3.03 2.0470 1.1973 78
Lack of Confidence 6 2.60 1.7137 .8359 78

N I Chi-Square I df I Sig.

78 I 61.014 I 5 I .000

 

 

Research Question 4: Predicting Participation Using Deterrent Factors
Correlation of Factors and Participation
Research Question #4 called for the ability to predict participation behavior based on the
deterrent factors identified. To initially examine the relationships between the six factors
and participation behavior, a Pearson r Correlation was performed between the two
variable types. Four of the six deterrent factors demonstrated a significant correlation
with level of participation (1 = nonparticipant; 2 = participant). The four factors were
Lack of Course Relevance, Lack of Confidence, Time Constraints, and Low Personal

Priority. Factors #4 and #5 did not show a significant correlation (Table 34).

139

Factor 1, Lack of Course Relevance and level of participation showed a low, positive
association of r = .163 (p < .01), indicating that course relevance was more important for
individuals who normally participate in educational programs. The correlation between
Factor 2, Lack of Confidence and level of participation revealed a low negative
association of r = -. 149 (p < .01), indicating that lack of confidence was a more important
deterrent for non-participants than for participants. Factor 3, Time Constraints
demonstrated a low, positive association with level of participation at r = .197 (p < .01),
which indicated that those who normally participate see time constraints as more of a
deterrent than non-participants. The fourth fictor, Low Personal Priority yielded a low,
negative association with level of participation at r = -.148 ( p < .01). This correlation
indicates non-participants are more apt to be deterred from participating by low personal

priorities than those who normally participate.

Table 34. Correlations Between Deterrent Factors and Level of Participation

 

Deterrent Factors
1 2 3 4 5 6
Lack of Lack of Time Low Scheduling Cost
Course Confl- Con- Personal Constraints
Relevance deuce straints Priority
Level of
Participation .163" -.149"”" .197" -.148** .067 -.056

 

*p < .05 ”p < .01

140

Logistic Regression for Predicting Participation Behavior

A logistic regression was performed to examine the ability to predictive participation
behavior for participants and non-participants based on the six deterrent factors. Using
the Chi-Square statistic, the analysis revealed that there was no significant relationship
between the observed and predicted for non-participants due to the deterrent factors.
While predictability is very high for the participant group with a value of 98.11 percent,
predictability for non-participants is very low at 11.54 percent (Table 35) and thus, the
results were not significant for the test. According to the results, the 6 deterrent factors
identified in the study could be used to predict participation behavior by those who
normally participate, but not for those who are non-participants. The results suggest that
perhaps there are difi‘erences between the two populations which are not explained by the
deterrents as they are now identified, yet impact their decisions not to participate in

educational programs.

To further explore these results, separate Pearson r correlations were performed for
participants and non-participants respectively, to examine the relationships between the
deterrent factors and sociodemographic variables. Table 36 shows the results of the
correlation for participants, while Table 37 shows the results of the correlation for non-

participarrts (Appendix F & G).

Results of the two correlations did show differences in how sociodemographic and

141

Table 35. Logistic Regression with Chi-Square Statistic: Test for Prediction of
Participation Behavior for Participants and N on-Participants in

 

 

 

 

 

Educational Activities.
Chi-Square df Sig.
Goodness-of-Fit 2.6339 8 .9552
PREDICTED
PERCENT
OBSERVED Non-Participants Participants CORRECT
Non-Participants 9 69 1 1.54%
Participants 5 260 98.1 1%
Overall 78.43%
VARIABLES IN THE EQUATION

VARIABLES B S.E. Wald df sig R Expo!)
Lack of -.4595 .2332 3.8834 1 .0488 -.0716 .6316
Confidence
Cost -. 1894 .1412 1.7990 1 .1798 .0000 .8274
Low Personal -.43 54 .2083 4.3715 1 .0365 -.0803 .6470
Priority
Lack of Course .3520 .2020 3.0372 1 .0814 .0531 1.4220
Relevance
Scheduling .1023 .1809 .3198 1 .5717 .0000 1.1077
Constraints
Time Constraints .6737 .1981 11.5697 1 .0007 .1613 1.9615
Constant .2479 .4910 .2548 1 .6137

 

142

deterrent variables relate for participants and non-participants based on the deterrent
factors identified. Significant relationships appeared in 10 correlations for the participant
group, while only 4 correlations emerged for the non-participant group. Two of these
correlations occurred between the same deterrent fictor and sociodemographic variable,
i.e., 1) Scheduling Conflicts and ofi-farm job and 2) Lack of Confidence and age. The
association between schedule conflicts and ofi-firm job was the strongest correlation
(moderate association) for both participant and non-participant correlations and was also
the strongest association for “all respondents” (Table 14). The correlation between lack
of confidence and age is logical and is supported by the literature (Johnstone and Rivera,
1965). An expected association between lack of confidence and level of education was
only observed in the participant group and did not appear for non-participants at a
significant level. An expected association between level of education and low personal
priority surfaced at a significant level for non-participants but not participants. Also the
deterrent factor Cost correlated only with gender for non-participants and not for those
who are in the participant group. Studies have shown that women typically cite cost more
strongly as a deterrent than men (Cross, 1965; Darkenwald and Valentine, 1985; Davis,

1988; Weischadle, 1988).

When comparing the findings of correlations of participants and non-participants
((Appendix F, Table 36 & Appendix G, Table 37) with all respondents (Table 14),
difi’erences appear regarding the correlations. This appears to support the supposition that

these two populations may difi’er in ways yet to be discovered regarding participation

143

behavior, therefore leading to an inability to predict participation behavior for non-

participants.

Open-Ended Items
Two open-ended items were placed at the end of the DPS-F questionnaire. The first of
these questions was stated, “Finally, please list the item that is most important to you in
your decision to participate in an educational activity.” This item was stated from a
positive perspective rather than negative as in the 39 deterrent items, in order to examine
(non-statistically) how the responses would compare with the results of the deterrent

items. These responses are listed in Appendix H, and support the findings of the study.

The final question was stated, “Feel flee to list below, any barriers to your participation in
educational activities that were not mentioned in this questionnaire? Also list any other
suggestions you have.” This item was not intended for comparison with the findings. It
was placed on the questionnaire in order for respondents to provide feedback about the
questionnaire, to share thoughts regarding the study, or to provide any other input. These

comments were retained by the investigator and not placed in the appendix.

CHAPTER 5

CONCLUSIONS, IMPLICATIONS, AND RECOMMENDATIONS
Conclusions

Introduction
The Conclusions, Implications, and Recommendations section explains and connects the
data with regard to findings in this study and in related studies. Implications of the
findings and recommendations for firrther study are also included. Sub—chapters and
sections include the Conclusions, Introduction, Deterrent to Participation Scale for
Farmers, Significance and Relevance of Deterrent Factors, Factors and Sociodemographic
Variables, Factor 1: Lack of Course Relevance, Factor 2: Lack of Confidence, Factor 3:
Time Constraints, Factor 4: Low Personal Priority, Factor 5: Scheduling Constraints,
Factor 6: Cost, Ranking Cost as a Deterrent, Predicting Participation Based on Deterrent
Factors, Open-Ended Responses, Implications of the Findings, and Recommendations for

Future Study.

The first objective of the study was to develop a valid and reliable Deterrent to
Participation Scale for Farmers (DPS-F) which could identify relevant principal

components (factors) to best represent the many deterrent items reported by farmers.

144

145

The second objective consisted of four research questions which would 1) test whether the
six deterrent factors were significantly difi’erent, 2) test whether there were difi’erences in
the type and level of the six deterrent factors based on sociodemographic variables, 3)
examine the importance of Cost as a deterrent, and 4) test whether participation behavior

can be predicted based on the deterrent factors identified.

Deterrent to Participation Scale for Farmers

The first objective of the study was performed to develop a valid and reliable DPS
instrument for firmers. Of the 1207 questionnaires mailed, a relatively high percent (41.9
percent) of the potential, reachable audience responded providing much useable data. The
reliability of the questionnaire remained high throughout implementation and resulted in an

alpha score of.9012.

Statistical analysis including Orthogonal Rotation using the Varimax Procedure revealed a
set of six source variables (fictors) which represented 50.523 percent of the explained
variance and was the most parsimonious solution in representing firmer deterrents to
participation. This percent of explained variance is within the range of values found in

other DPS studies, which again supports the DPS-F as a valid and reliable instrument.

Because scale items represented a negative perspective, it was appropriate to use factor
labels that also represented a negative perspective. The factors that emerged closely

resembled deterrent fictors identified in previous DPS studies, although in a difl‘erent

146

order of importance based on explained variance. In fict, the results of the Orthogonal
Rotation process showed that many of the deterrent items remained in the same factor
groups as in past DPS studies. As a result, fictor labels from previous DPS studies were
used to identify the DPS-F factors. The six fictors, in order of importance of explained
variance, were labeled Lack of Course Relevance, Lack of Confidence, Time Constraints,
Low Personal Priority, Scheduling Constraints, and Cost. Total item mean scores for the
39 Likert scale items tended to be low, ranging fiom 3.5 (“Somewhat Important” to
“Quite Important”) for the highest ranking item to 1.32 (“Slightly Important”) for the
lowest making item. As shown in earlier DPS studies, consistent low mean scores on scale
items support the position that the deterrent construct is multidimensional and not
dependent on a single type of deterrent. The ten highest ranked items by mean score for
the entire scale are classified as situational and institutional, which compares with the
typology of Cross (1981) and with other DPS findings. This further confirms the DPS-F

as a valid instrument to measure deterrents to participation for this subgroup.

The DPS-F performed well in identifying deterrent fictors for firmers and can be used as
a valid and reliable instrument in identifying farmer deterrents to participation. Thus, the

first objective of this study was achieved.

Significance and Relevance of Deterrent Factors
The first null hypothesis examined whether the six deterrent fictors were significantly

distinct and separate. Null Hypothesis #1 was stated, “There are no statistically significant

147

and distinct deterrent factors to explain farmer participation in educational activities.” To
address this question, the six factors identified through Orthogonal Rotation were
transformed into new variables utilizing total item mean scores within each factor. The six
new variables were then compared using a repeated measures test with multivariate
analysis of variance and paired t-tests. Results showed that the six factor variables were
indeed significantly different at the p < .001 level, and represented distinct and separate
source variables that could be utilized further to analyze sociodemographic and
participation data. Therefore, based on the repeated measures test and F—statistic at the p

< .05 level, Null Hypothesis #1 is rejected.

Factors and Sociodemographic Variables

Null Hypothesis #2 was stated, “There is no significant difi‘erence in firmers’ deterrents to
participation in educational activities based on sociodemographic variables such as age,
level of education, amount of gross farm sales, farming status, or those holding or not
holding an ofl’-firm job.” This question examined whether firmers from difierem
sociodemographic categories difi‘er regarding the type and importance of the six identified
deterrent factors that serve as barriers to participation in educational activities. No
attempt was made to combine effects of sociodemographic variables, e.g., age and income

in regard to a deterrent factor.

To address this question, a Pearson Product-Moment Correlation analysis was

implemented to initially compare factors and sociodemographic variables for significant

 

148

relationships. Farm size, by both acreage and number of square feet, was not analyzed due
to the wide range in farm sizes reported, and because firm size has little meaning without
linking it to firm type. Of 36 possible correlations between the six deterrent factors and
six sociodemographic variables, the analysis revealed 16 correlations that emerged at a
significant level of p < .05. Of the 16 significant correlations, 15 showed only “low”
association, i.e., r = .20 to .30 (Elliot, 1991). The remaining one correlation between
Factor 5, Scheduling Constraints and the sociodemographic variable, Off-F arm Job

revealed a “moderate” association , i.e., r = .30 to .49 (Elliot, 1991).

In order to test the hypothesis regarding sociodemographics and the deterrent fictors, a
One—Way Analysis of Variance was performed where each significant correlation had
appeared. In two of the ANOVA tests, no statistical difi‘erence was found, i.e, between
Factor 1, Lack of Course Relevance and annual gross firm sales, and between Factor 3,
Time Constraints and age group. The variables Time and primary enterprise type were not
analyzed using a Pearson Correlation, however an ANOVA with Tukey HSD was used to

look for significant difl’erences due to enterprise type.

Factor 1: Lack of Course Relevance

The source variable (Factor 1) capturing the highest amount of explained variance in the
rotation was labeled Lack of Course Relevance. This factor correlated significantly with
annual gross sales only, indicating that the importance of course relevance increases with

income level (annual gross sales). However, this correlation was weak. Likewise, the

149

ANOVA did not reveal a significant nor meaningful difference between the three income
groups. This does not mean that course relevance is unimportant. In fact, the fictor
scale mean surfaced as being second-highest of the six factors with a value of 2.47, i.e.,
“slightly important” to “somewhat important”, while the mean for the entire scale
instrument was 2. 53. The data appear to show that this item is a relatively important
deterrent among all farmers, but that it does not seem to vary due to sociodemographic
group. This factor resembled previous deterrent research findings in that all items within

this fictor are of either the institutional or situational type.

Factor 2: Lack of Confidence

Lack of Confidence correlated significantly with age, level of education and holding an
off-farm job. The relationships between lack of confidence and these independent
variables are logical and agree with the literature (Johnstone and Rivera, 1965;
Darkenwald and Merriam, 1982). The data indicated that lack of confidence increases
with age, particularly with farmers 65 years of age and over. Results of the analysis of
variance indicate that there is a significant and meaningful difi‘erence between those who
are 65 and over and those in the 18-34 age group, and between those who are 65 and over
and the 45-54 age group. According to the data, firmers in the oldest group are more
likely than the other two groups to avoid participation due to lack of confidence. The
study also showed that lack of confidence decreases with increasing levels of education
and for those working 03 the firm, however difi‘erences did not vary enough in the study

between levels of education nor between those holding ofl’-farm jobs and not holding ofil

150

farm jobs to be meaningful.

Factor 3: Time Constraints

Factor 3, Time Constraints correlated significantly with age, annual gross sales and
farming status, indicating that time constraints become less of a deterrent with age, but
more of a deterrent with increasing farm sales (income) and those firming firll-tirne versus

part-time.

The relationship between time constraints and age appears to be a logical pattern with the
largest variation in mean scores occurring between farmers age 35-44 and firmers 65 and
over. This could be a practical observation, i.e., the age group 35-44 is likely about the
age that farmers are assuming ownership of the family firm as the fither enters retirement.
However, the analysis of variance demonstrated that no significant nor meaningful

difference occurred between these two variables.

Time constraints would also likely become more important as gross farm sales rise, since
this implies more product sold and likely higher time requirements in managing the farm.
Significance was seen between the low gross farm sales group ($9,999 or less) and high
gross farm sales group ($100,000 or more), and also between the medium gross firm sales
group ($10,000 to $99,999) and high gross farm sales group. Although time constraints
provided a continuum of increasing importance with increasing farm sales, there was no

significant difference shown between the low and medium gross firm sales groups, nor did

151

a meaningful difl‘erence occur between the middle and higher gross sales groups. The data
do show, however, that a significant and meaningful difference does exist between the
lower gross firm sales group and upper gross farm sales group. This is a logical result,
since the data show increasing importance of time as a deterrent as farm sales rises.

F arrners in the higher gross firm sales group appear to have much more difiiculty in

finding time to participate than firmers in the lower gross sales groups.

The data also show that a significant difl'erence occurs between full-time and part-time
farmers suggesting that full-time farmers have more time constraints than part-time
firmers. However, the difl’erence in mean scores from the study are not suficient to

support a meaningful difl’erence due to firming status.

Finally, the ANOVA and Tukey HSD tests between Time and primary enterprise showed
that a significant and meaningful difference did appear in four analyses. Time constraints
were shown to be significantly more important as a deterrent 1) for dairy farmers than for
cash crop producers, 2) for dairy firmers than for livestock producers, 3) for dairy farmers
than fiuit growers, and 4) for diary farmers than finners in the “other” enterprise
category. While mean scores between dairy and equine producers did equate to more than
a .5 interval, significance appeared at p = .150 and therefore was not included as being
significant nor meaningful. The low number of equine producers (n = 25) may have led to

this outcome.

152

Factor 4: Low Personal Priority

Factor 4 dealt with the farmer’s disposition toward learning and a priority of how that
person wished to use available time. Because a negative label was necessary to describe
this barrier type, “Low Personal Priority” was chosen. Factor 4 showed significant but
low negative associations with gender, level of education, annual gross sales and firming
status, indicating that a person’s priority to learn at other times or in other ways decreases
as a deterrent as education levels increase, as gross sales increase, and in moving fiom firll-

time to part-time firming status.

According to the results of this study, females place less importance on how and when
they learn than males, which was a similar finding to Darkenwald and Valentine (1985) in
the DPS study with general audiences. However, the results of the ANOVA show tint
there were not sufficient difi‘erences in the two mean scores to suggest a meaningful

difi’erence.

The relationship between low personal priority and level of education is supported in the
literature by Houle (1961), Cross (1981), Darkenwald and Merriam (1982), and also by
Weischadle (1988) in a study of real estate professions. These studies show that adults
with low levels of education tend to place a lower priority on continuing education than
do persons with higher levels of education. This trend is also seen in mean scores of
firmers in the four levels of education, however insuflicient ranges between the lowest and

highest mean scores indicate that a meaningful difl’erence does not exist between the

153

education levels based on the ANOVA tests in this study.

Low Personal Priority and farm gross sales, while showing a significant correlation, did
not show significance during AN OVA tests nor did any meaningful difl’erences emerge.
Low Personal Priority did result in a significant relationship with farming status, indicating
that those who firm full-time based less importance on this variable as a deterrent than did
part-time producers. However, in AN OVA tests, the variation in mean scores did not

produce a meaningful difference due to firming status.

Factor 5: Scheduling Constraints

Factor 5, Scheduling Constraints correlated significantly with age, annual gross sales,
farnring status, and on- or ofilfirm employment, indicating that scheduling constraints
become more of a deterrent with those holding an off-firm job versus no off-farm job, but
less of a deterrent as firmers grow older, as sales increase, and as firmers move fiom part-
time to full-time. The correlation between scheduling constraints and those who hold ofl‘-
farm jobs was the strongest (“moderate”) association of the 16 correlations identified.
This was further supported by correlations performed separately under Research Question
#4 (Appendix F), where moderate associations emerged for both participant and non-
participants, respectively between these variables. This appears to indicate that one of the
more important deterrents to participation for firmers exists in scheduling conflicts for

those firmers who hold an ofi’-farm job.

154

When the null hypothesis was tested using AN OVA, variation in mean scores for all
variables except holding an ofi-firm job resulted in less than a meaningful difference in
mean scores. However, the AN OVA did indicate a meaningful difference in Scheduling
Constraints as a deterrent between those holding or not holding an off-firm job. Those
holding ofilfarm jobs appear to have more difficulty in balancing schedules to participate

in educational activities than firmers without ofi-farm jobs.

Factor 6: Cost

Factor 6, Cost correlated significantly only with gender, showing that females see cost as
more of a deterrent than do males. This is supported in the findings of Cross (1981),
Darkenwald and Merriam (1982) and in DPS studies by both Davis (1988) and
Weischadle (1988). It is interesting to note, however, that the correlations performed in
relation to Research Question #4, i.e., for participants and nonparticiparrts, respectively, a
significant relationship between these two variables emerged only for non-participants. In
fict, a moderate association of r = .341, appeared. The correlation between these two
variables for farmers in the participant group did not reveal a significant association
indicating that cost may be more important for females who do not typically attend
educational activities. Nevertheless, for all respondents as a group, cost does serve as a
significant and meaningful deterrent to participation for females, based on the results of

this study.

An ANOVA between Cost and gender did reveal that a significant and meaningful

155

difference exists. Female farmers indicate that cost is more of deterrent to their

participation in educational activities than do males.

In summary, five significant and meaningful differences were seen between the six
deterrent fictors and seven sociodemographic variables. Lack of confidence was more of
a deterrent for firmers over age 65 than for farmers in the 18-34 and 45-54 age groups.
Time constraints were more of a deterrent for firmers in the highest gross firm sales
category than for firmers in the lowest gross firm sales category. Schedule constraints
were more of a deterrent for firmers who hold ofi-firm jobs than for those who did not
have an ofi-firm job. Cost was more of deterrent to participation for female farmers than
for male farmers. Time constraints were a more important deterrent for dairy producers
than for cash crop producers, livestock producers, fiuit growers, or farmers in the “other”

enterprise category.

The results of the study indicate that significant and meaningful difl‘erences are shown in
regard to the deterrent factors as the sociodemographic groups vary. Based on results of
the One-Way Analysis of Variance and Tukey HSD Tests using the F—statistic at the p <

.05 level, Null Hypothesis # 2 is rejected.

Ranking Cost as a Deterrent Factor
The third null hypothesis focused on the deterrent factor labeled “Cost” to determine if

this was the most important deterrent for farmers in their decisions to participate in

156

educational programs. Null Hypothesis #3 was stated, “Cost is not the most statistically
important deterrent in farmers’ decisions to participate or not participate in an educational

activity.”

To address this question, a Friedman’s Non-Parametric Test was implemented to
statistically rank the six deterrents in order of importance. This was performed for all
respondents as a group, and also for non-participants as a group in a separate test to see if
the two groups differed in how the fictors ranked. Results of this test revealed that
regardless of the group selected, the deterrent factor Cost emerged as the fifth most
important fictor among the six deterrents identified for firmer-s. Time Constraints and
Scheduling Constraints emerged as the most important and second most important
deterrent, respectively, for the total respondent group and for the non-respondent group.
Lack of Course Relevance surfaced as the third most important deterrent for all
respondents as a group, with Low Personal Priority emerging as fourth most important.
Lack of Course Relevance and Low Personal Priority were in reverse order for the non-
respondent group, i.e, Low Personal Priority (third) and Lack of Course Relevance
(fourth). Both groups indicated Lack of Confidence as the least important deterrent,

based on the six fictors identified.

Therefore, since Cost ranked as the fifth most important deterrent in both Friedman tests,
it is shown that Cost is not the most important deterrent for farmers in their decision to

participate in an educational activity. Thus, based on results of Friedman’s

157

Non-Parametric Test using the Chi-Square statistic at the p < .05 level, Null Hypothesis

#3 is accepted.

Predicting Participation Based on Deterrent Factors

The forth null hypothesis examined whether participation behavior could be predicted with
respect to the sociodemographic traits and the deterrent fictors. Prior to testing the null
hypothesis, a Pearson r correlation was performed to identify significant associations
between participation behavior and the deterrent factors. The Pearson correlation
revealed significant associations on four of the six fictors indicating that relationships did
exist with Lack of Course Relevance, Lack of Confidence, Time Constraints and Low
Personal Priority. Preliminary results of these correlations indicated that Lack of Course
Relevance and Time Constraints were more important as a deterrent for participants than
non-participants, and that Lack of Confidence and Low Personal Priority were more

important deterrents for non-participants than participants.

To test the null hypothesis, a Logistic Regression was performed. While the Logistic
Regression was able to predict participation behavior 98 percent of the time for the
participant group, it demonstrated a very low success rate in predicting for non-
participation (11%). These findings indicate that the DPS-F, while usefirl in identifying
farmers’ deterrents to participation, cannot be used to adequately predict participation for
the non-participant group. The results suggest that the non-participant group differs in

some way that could not be revealed through the deterrent factors identified. Therefore,

158

based on the Logistic Regression using the Chi-Square statistic at p < .05, Null Hypothesis

#4 is accepted.

Because the DPS-F was shown not to be useful in predicting participation behavior of the
‘hon-participant” farmer group, firrther exploration was performed to see if participants
and non-participants varied with respect to participation behavior and the deterrent
fictors. Pearson r correlations were performed separately on both the participant group
and non-participant group, using the sociodemographic variables and the deterrent fictors.
Results of the correlation demonstrated that significant associations occurred in 10
correlations for participants and only 4 correlations for non-participants (Appendix F,
Table 36 & Appendix G, Table 37). Only two of these correlations involved the same two
variable combinations for both participants and non-participants. Scheduling Constraints
and ofl‘-farm job resulted in a moderate positive association for both participants and non-
participants. Lack of Confidence and age resulted in a low positive association for both
groups. Also, Cost and gender resulted in a positive, moderate correlation, indicating that
for non-participants, Cost is a more important deterrent for females than for males. The
findings of the correlations indicated that further study is needed to examine how
participants and non-participants vary, in regard to participation behavior. The results
suggest that perhaps there are difl‘erences between the two populations which are not

explained by the deterrents as they are now identified.

159

Open-Ended Response

Respondents were provided opportunities to provide feedback through two open-ended
items placed at the end of the questionnaire. The responses from the first item allowed
farmers to indicate what was most important to them in their decision to participate in an
educational program. Responses were diverse ranging fi'om time and location, to the
relevancy of the information to their needs (Appendix H). Farmers also commented that
they were too old to continue their education, or were not aware of educational activities
being available to them. Some responses were very specific, e.g., relating to a very
specific topic that would spur their interest enough to attend, while others stated more
general items. The diversity ofthe responses appeared to support the findings in the
study, i.e., that the deterrent construct is not comprised of a single type of barrier, but that

it is complex and multidimensional in nature.

Responses fiom the second item included such comments as suggestions for additional
deterrent items that were not included in the study, to notes of thanks for being selected to
participate in the mail survey. As expected, because of the apparent (and intended) source
of the survey, comments also appeared in relation to Michigan State University and its

programs — both constructive and complimentary.

Implications of The Study
The development of a Deterrent to Participation Scale for Farmers (DPS-F) provides the

first known instrument of its type to identify and measure deterrents to participation

160

specifically for firmers. The study demonstrates that it is a valid and reliable tool for both
researchers and other educators to use in identifying and analyzing deterrents to
participation. The instrument can be used to examine a general population of farmers, or a
specific producer group. Examining deterrents to a specific producer group will be more
valuable when attempting to plan educational activities for that particular type of firmer.

Those barriers can then be addressed to increase grower participation.

While Extension educators were often mentioned in the study as important players in
transferring educational information to farmers, there are many other public organizations
and private firms which ofl’er educational programs and tours for firmers. The DPS-F and
it’s findings should be useful for all of these organizations to help them better understand
their audiences and customize programs to reduce or avoid deterrents to firmer

participation.

The study also underscores the importance of “deterrents” in the educational arena.
Motivational research has shown that adults participate in learning experiences for certain
reasons, and that these reasons vary from individual to individual. These studies have also
shown that similarities exist with respect to sociodemographic group. Research has
shown that deterrents also vary from individual to individual and with respect to
sociodemographic group. Therefore, educators must consider both the motivational

incentives and the barriers when planning and implementing educational activities.

161

Advisory groups or committees that work to plan programs ofien focus on those issues
which meet the needs of the target audience. Discussions on program content, potential
resource persons, program date and location, etc, are all fitting topics when planning
programs. However, advisory groups should remember that these program components
can serve as deterrents as much as incentives if not handled with the target audience in

mind

In addition to identifying six major deterrents that impact Michigan firmers, the study was
able to rank the deterrents according to importance as individual deterrent fictors. The
rankings were almost identical for both the total sample and the non-participant group.
Time constraints and scheduling constraints appear to be the two leading deterrents for
farmers and these compare fivorably with previous research findings for adult learners

(Johnstone and Rivera, 1965; Cross, 1981).

Lack of confidence, while tending to be a larger deterrent for certain sociodemographic
groups, ranked last for the entire group sampled in this study and therefore may not be as
important for general firmer audiences. However, the results of this study were similar to
previous studies which indicated that lack of confidence can be more of a deterrent to
participation for older adults tlmn for younger adults. While some older finners indicated
that they were “too old to learn”, the agricultural arena is one where older adults still have
an important role. Unlike many other types of industries, firmers ofien continue to provide

important input into firm management decisions well afler the age of 65, especially on

162

family owned and operated farms. Therefore, firmers over 65 years of age are still an
important target audience for organizations which provide educational programs for
farmers. Programs targeted at older farmers may need to consider promotional techniques
that would increase audience confidence levels. Educational activities that target general
audiences, such as using computers on the farm, could be ofi’ered in difl'erent formats with
one format ofl‘ered to younger farmers who may have more computer experience or more
confidence in learning, and in a difi'erent format for older farmers who may have a “fear”

of computer technology.

Cost did not surface as the most important deterrent among those factors identified.
Therefore, it may not serve as a barrier for those agricultural producers who have the time
to participate and are truly interested in the program content. This finding supports the
results of a firm financial workshop held at Michigan State University where participants
were asked to pay a substantial registration fee. The topic of the workshop was relevant
to farmers’ needs at that time and apparently considered other items that could have
served as barriers. As a result, the workshop was filled to capacity with 51 producers
attending and revenues ofi‘-setting some, or all, program implementation costs. However
based on the findings in this study, if registration fees are charged, the activity must be
relevant to the information the firmer needs, and ofl‘ered at time or season when

attendance is convenient.

In addition, the findings showed that females, as a specific sociodemographic group, do

163

identify cost as a deterrent. This appears to be especially true for female farmers who do
not usually participate in educational activities. This may be a direct result of the role that
women often play on the firm, i.e., the bookkeeper. On many firms, the farm wife
handles the financial records. It is often said that the person who “holds the purse-strings”
is more cautious of how money is spent. Regardless, organizations ofl’ering programs
where women are the target audience or a portion of the target audience must consider

cost as a potential barrier and plan accordingly.

The study indicated that time constraints are significantly more important as a deterrent to
participation for dairy producers than for other types of producers, such as cash crop and
livestock producers, fiuit growers, or farmers representing the “other” category in the
study. This is an important finding since agricultural programs for specific sub-groups are
often sub-group specific. For example, programs addressing such topics as dairy calf
morality, milk quality, and sire selection would not be pertinent to most other types of
firmers other than dairy producers. It is well known that most dairy producers milk cows
and do related chores in the early morning and late afternoon. Extension agents realize
this and typically start educational programs for dairy firmers later in the morning and end
the programs in time for producers to return to the firm for the evening milking. The
results of this study reveal however, that time constraints may be an even larger barrier

than previously thought for dairy farmers, and additional study may be warranted.

The DPS-F instrument was not successful in predicting participation behavior for non-

164

participants, although it did provide a very high level of prediction for those who normally
participate in educational programs. It appears, from follow-up correlations in this study,
that there may be important difl‘erences between farmer “participant” groups and “non-
participant” groups. Additional research must be performed to determine whether there
are other unknown variables at work which impact participation decisions by non-
participants. This may mean that the two groups should be studied separately, or perhaps,
a modified DPS-F for non-participant types should be developed through focus groups
consisting of only non-participant farmers. While it is important to concentrate on non-
participants and how educators might find ways to attract those individuals, deterrents
also exist for those who do participate. Even individuals who normally participate in
educational activities cannot attend every learning activity offered. It will be important to
continue investigations of both participants and non-participants and the barriers which

impact their participation.

Comparing the DPS-F findings with earlier deterrent research confirms that differences do
vary with populations or subgroups. We cannot simply assume that one group’s
deterrents, such as those of the general audience, will be similar to another group’s
deterrents. Therefore, additional study of specific sub-groups will lend much to our

understanding of barriers to participation in educational programs.

The study confirms the importance of knowing the target audience when plamring

educational programs. The issue of deterrents to participation in educational programs,

165

while important for outreach units such as Cooperative Extension Service, is just as
critical for the university general which administers the Extension program. Universities
reach out in many ways to adult audiences, through on-going or special programs,
institutes, on-campus and ofi-campus course work, etc. The success of the DPS
instruments in identifying deterrents for distinct populations demonstrates that it can also
be modified for use with other target audiences of the university. The DPS-F findings
indicate that cost is not the most important deterrent to farmers in Michigan, and that
program fees could be considered as a way to recoup expenses. This could have positive
implications for Extension in Michigan - as well as other programs and institutes at
Michigan State University which target farmers. Additional DPS studies of other MSU
target audiences may show that this is true for other outreach programs and audiences as

well.

Reconmrendations
This study was intended to further expand the findings and foundation of the deterrent
construct by examining a population not yet investigated in previous DPS studies. The
results of the Deterrent to Participation Scale for Farmers (DPS-F) investigation support
the construct initially proposed by Darkenwald and Scanlan (1984), in that the construct

appears to be sophisticated and complex, and represents a multidimensional premise.

This study provides the first known instrument of its type to examine deterrents to

participation for firmers. The DPS-F study was successful in identifying and ranking the

166

deterrents indicated by Michigan farmers. Because the DPS-F instrument was able to
identify statistically distinct deterrents for agricultural producers and provide a ranking of
importance of those deterrents, it is recommended that further exploration take place with
other sub-populations such as fiuit growers, dairy producers, young farmers, females, or
members of the firm fimily who keep the farm financial records, to see how these
audiences might be better targeted when planning educational programs. Although we
know the rankings in importance of the six deterrents in this study, they exist for Michigan
farmers in general and for non-participant types. The rankings of these deterrents may

also vary by sociodemographic group.

The application of the DPS-F as a reliable and valid instrument to predict participation did
not prove useful for non-participants. However, the high level of prediction for firmers
that normally participate implies that it can be a reliable and valid tool. Additional study is
needed to determine ways in which to enhance prediction ability of non-participant firmers
using the DPS-F. The inability to predict participation behavior for non-participants may
demonstrate that non-participant audiences need to be analyzed separately. It is not clear
fiom the results of this study whether other unknown variables are impacting decisions to
participate by non—participants. One method to address this issue is to develop a modified
form of the DPS-F instrument, where deterrent scale items are developed by non-
participant types only. Therefore, it is recommended that the instrument be tested solely
on a non-participant firmer sample to check for predictability with that group, and to

compare, generally, how the deterrents for non-participants compare and contrast with

167

deterrents for the participant group. Other DPS studies have occasionally incorporated
complementary scales, e.g., the Adult Attitudes Toward continuing Education Scale
(AACES), with the DPS instrument while others have incorporated job roles (W eischadle,
1988; Davis, 1988). The results have been mixed in regard to predictability, however this

may assist in strengthening the ability to predict participation in firm audiences.

This particular study addresses deterrents to participation in educational programs for
Michigan farmers with particular attention to various sociodemographic groups. Much
additional study is needed to better understand the deterrents that afi‘ect Michigan farmers,
and how these vary between specific groups. Additional study of various commodity
groups would be valuable for Extension in addressing the farmer’s educational needs.
Also, follow-up DPS-F surveys of sociodemographic groups based on age, gender, level
of education, geographic area, or cultural group, etc, will assist in qualifying the findings
of this study, while helping educators in this state to better understand Michigan firmers
Because Michigan farmers consist of both part-time and full-time producers, it would help
greatly to better understand the differences in motivators and barriers for these two

groups.

The deterrents identified in this study may or may not represent firmers in other
geographic areas. Farmers in California or South Dakota, for example, may have very
difl’erent deterrents due to weather impacts, geographic distances between firms and

towns, availability of quality educators, or accessibility to educational activities based on

168

budgets (limitations or surpluses) of Extension or other agricultural organizations which
provide educational activities. Statewide DPS-F surveys, or surveys of specific firm
groups should help Extension and other educators learn more about their potential

audiences regardless of the state or region being explored.

While cost was shown to be less important than other deterrents, further investigation is
warranted on this issue. Cost does appear to be more of a deterrent for female farmers
than male farmers, which supports previous studies dealing with barriers to participation
for adults in general. At this time, we do not know if this is due to the fict that the firm
wife is often the firm business bookkeeper, i.e, more prudent in spending farm fimds, or if
there are other reasons. The correlations between the deterrent factors and participation
behavior (Appendix F & G) show that there is no significant relationship for Cost between
males and females for those classified as ‘participants”. However, a moderate association
appears for the “non-participant” group, showing that Cost is a significantly more
important deterrent for females than males. Based on this finding, the entire issue of Cost,
with regard to relationships and difl’erences between male and female farmers, needs

additional study.

The Cost factor has shown up in several deterrent studies, but not all studies compare the
importance of this factor with other deterrent fictors. The DPS-F study has shown that it
is not the most important barrier for firmers, and that other issues such as time constraints

and schedule conflicts, program relevance, and low personal priority at the time of the

169

program offering, may outweigh cost as a deterrent. Additional study of firmers’
behavior in response to registration fees and other program costs, would help social
scientists and economists to better understand how firmers react to Cost as a potential
deterrent. The DPS-F study implies that if the program is ofi‘ered at a convenient time,
day or season, and includes information needed at that time by firmer clientele, they will
participate regardless of costs. This finding is not unexpected and is likely a cost-benefit
issue. In this case, the value is placed on the returns of participating in an educational
activity. Due to tighter budgets for public education institutions, including Cooperative
Extension Service and institutions of higher learning in general, it is important to know
when clientele are willing to pay a fee for the privilege of participating and when they are
not. If there are ways to recoup expenses through program fees, then many educational
institutions could benefit financially. The DPS methodology should be considered to
explore this arena by utilizing DPS instruments already in place, through modifications of
existing DPS templates, or by developing new DPS instruments through focus group
input. It may also be possible to incorporate methodologies that include the theory of
value, such as contingent valuation, to see how those constructs complement the deterrent

construct in educational research.

The findings in this study support not only the deterrent construct, but other facets of
educational research such as program planning techniques. Barriers such as time
constraints, schedule conflicts, relevancy of program topics or level of learning are all

components that a good program planner should take into account, regardless of the

170

audience. Extension staff and other individuals and organizations who target farm
audiences will want to explore those audiences with respect to the deterrents that may
keep them from attending. According to the DPS-F findings, the quality of an educational
program should not be sacrificed due to the cost of implementation, however female
participants may find cost as a deterrent more than males. The results show that farmers
in general may be willing to pay registration fees, if other deterrents can be excluded.
Extension agents, typically guided by limited operating budgets when planning programs,
may be able to provide enhanced quality to programs by recapturing costs or creating
revenues that could be returned to program implementation. Regardless, Extension agents
in Michigan , based on the findings in this study, should be more fearful of loosing
audience participation due to the timing of the activity, due to inadequacies in program

content, or due to the method of program delivery, than due to program costs.

One thing is certain. Technological advancement is impacting farmers and their practices.
This relates not only to operating in an environment of global market forces, but also to
breakthroughs in computer technology, genetic engineering, crop and animal
management, etc. Farmers have no choice but to participate in continuing education if
they wish to survive. At the same time, educators who provide information for firmers
must be able to deliver programs and activities in a manner that is acceptable to US.
farmers. Extension stafl‘ and other educators should place critical importance in
understanding participation behavior related to motivators and deterrents for their

clientele, and in regard to determining clientele needs. This underscores the importance of

171

advisory group input when planning educational programs.

Deterrents to participation do exist for farmers and they vary by population and
sociodemographic group. The DPS-F may help scientists and educators in the future, in
identifying those deterrents for other firmer types or in other firming areas. Regardless,
the study has added to the discipline of adult education and has taken us one step closer to

understanding deterrents which affect learning behavior of an adult population.

APPENDICES

APPENDIX A

UNIVERSITY COMMITTEE FOR
RESEARCH INVOLVING HUMAN SUBJECTS

 

517m
FAX: 51 71432-1171

mung-um
WOW”:
mum

lam-W

January 19. 1998

TO: Carroll H. Wamhoff
410 Agriculture Hall

FR: David E. Wright, Ph.D.
Chair
The university Committee on Research

Involving Human Subjects (UCRIHS)

246 Administration Building

RE: IRB#: 97-122
TITLE: DETERRENTS TO PARTICIPATION IN
EDUCATIONAL ACTIVITIES FOR MICHIGAN
FARMERS
CATEGORY: l-C

APPROVAL DATE: 02/21/97

Our records indicate that this project was .
approved on the date shown above. As you know,
UCRIHS approval is valid for one calendar year.
If you are planning to continue your study after
February 21, 1998, you must complete and return
to the UCRIHS office a green renewal application
form by January 21, 1997. There is a maximum of
four such expedited renewals possible.
Investigators wishing to continue a project beyond
that time need to submit it again for complete
revrew.

As you are aware, UCRIHS must review and approve
all revisions to human subjects activities, prior
to initiation of the change. Therefore, if you
have any future study revisions you wish UCRIHS to
review and approve at this time, please answer
question #7 on the renewal form 'no' and follow
the instructions given there.

If you have decided to discontinue the research or if you
have already submitted your application to renew this study,
please disregard this reminder.

THE APPLICATION FOR RENEWED APPROVAL HAS BEEN SENT TO THE
RESPONSIBLE PROJECT INVESTIGATOR (ADDRESSEE ABOVE).

cc: Mark F. Hansen

172

APPENDIX B

SURVEY INSTRUMENT

PARTICIPATION IN
EDUCATIONAL ACTIVITIES

A Questionnaire for Michigan Producers

 

AEE Center of Evaluative Studies
408 Agriculture Hall
Michigan State University
East Lansing, MI 48824-1039

173

174

 

PURPOSE OF THE STUDY
Thepurposeofthisstldyistohelptheagrlenlturalind ustrybu ‘ ‘ ‘ “‘ “ ‘
preventfarners'l'rom participating in educational activities. Your help in lileitifylng these

reasons, is very important to us. Would you please help us by completing this brief questionnaire.
Please respond as honestly and as thoughtfully as you can. You are m requested to put your
name on this survey.

 

 

The following example will help demonstrate how to complete this survey.
(Circle the number that best describes your opinion or situation .)

Even though I wanted to learn more about the topic covered, I decided not to attend a recent educational
activity...

 

This reason [or not participating was-
Reasons Not Slightly Somewhat Quite Very
lmpm-rsm Important important Important Important

 

1. Because of the location where it
was held .................... ®

By circling '4, " the respondent indicates that the lomtion of the educational activity was "quite
important" when deciding not to participate

Are you at least 18 years of age and actively involved in an agricultural or horticultural businas
attire present time? (Examples include alarm, orchard, nursery, etc.)

YES

If "NO“, please pass this questionnaire to an individual on your farm or business who meets these
criteria .

 

175

First of all, we recognize that people learn in different ways. Participation in planned and
organized educational activities is one method that people may use to learn new things. These
may be related to work or just taken for enjoyment.

How many educational activities did you participate in during the past year? Include both
agricultural and non-agricultural related educational activities. lnclude such things as educational
tours or meetings. lessons. conferences. demonstrations. etc.. (Circle an: answer only.)

None

l-S activities

6-10 activities

1 l or more aetivt'ties

"3 OJ e-fi

Ia

Think of an educational activity - whether farm-related or not - that you WANTED to attend
within the past 12 months. but never did. Examples might include flying lessons. computer
training. a marketing workshop. etc.

How important was each reason listed below. in your decision NOT to participate in that
educational activity '.' (Circle only one answer per question.)

 

This reason for not participating was...
Reasons Not Slightly Sou-wan Oule Very
W import-u Import-n Imam lutportam

l Because the activity was offered during a

 

 

busy time of year 1 2 3 4 5
2 Because the activity schedule conflicted with
my off-farm job I 2 1 4 5
3 Becausctherewasnooncatthe {armor
business who could cover for me .................. l 2 3 4 5
4 Because the farm or business comes first
and l didn‘t think I should be gone ................ l 2 3 4 5
5 Because there were too many activities
offered at the time to attend all of them .......... l 2 3 4 5
6 Because I wanted to learn something
specific but the activity was too general ........ l 2 ' 3 4 5

(Continued)

Reasons

7 Because pre-registration was required
and I didn‘t want to make a commitment
that far in advance

8 Because the activity that was offered did not
seem interesting enough

 

9 Because the activity was offered at an
inconvenient location

10 Because I couldn‘t afford the registration
or aetivity fees

1 1 Because I felt I was too old to participate
in the activity

 

12 Because I didn't know about educational
activities available on this topic

13 Because of the amount of time required
to attend all of the sessions

14 Because the activity was scheduled at

 

an inconvenient time of day

15 Because my family did not encourage
my participation

16 Because I prefer to learn through other

methods such as reading. radio, etc .....................

17 Because 1 thought that the activity available
would be of poor quality

 

18 Because 1 was not confident of my
learning ability

 

19 Because of conflicts with my children's
activities

176

 

This reason for not participating was...

Not
Important

Slightly
Important

L.)

'9

DJ

DJ

k)

id

'0

IO

L)

I»)

Somewhat
Important

b)

DJ

b)

Quite
Important

Vet)
lmportan:

‘Jt

177

(Continued)

 

This reason for not participating was...

Reasons Not Slightly Somewhat Oune \fl)
Importan: important Important Important Important

20 Because I'm not that interested in

 

 

 

 

 

 

 

 

 

attending educational activities ...................... 1 2 3 4 5
21 Because my participation would take
away time from my famiiy 1 2 3 4 _~
22 Because I didn‘t receive enough information
about the actinty ahead of time ............................ 1 2 3 4 5
23 Because the available actmty did not
seem useful or practical 1 2 3 - 5
24 Because 1 wasn‘t willing to give up my
leisure time l 2 3 4 s
25 Because a last-minute conflict arose that
prevented me from attending 1 2 3 4 5
26 Became I felt unprepared for the
activity 1 2 3 4 5
27 Because I couldn‘t afford miscellaneous
expenses like navel. meals. etc ......................... l 2 3 4 5
28 Bacausc the activity was not on the right
level for me 1 2 3 4 5
29 Because the aeuvity conflicted with my
civic or eommrniity duties 1 2 3 4 5
30 Because 1 didn‘t tlunk the activity would
meet my needs 1 2 3 4 5
31 Becauscldonoireadorwntewell enough
to feel comfortable particrpoting ........................ 1 2 3 4 5
32 Beausc I did not feel that the activrty was
worth the fee for ‘ ‘ g 1 2 3 4 s
33 Because 1 lacked the energy to attend ........... 1 2 3 4 s

178

(Continued)

 

This reason for not participating was.
Reasons Not Slightly Somewhat owe Very
important Important Important Important Important

3: Because the activity was not offered

 

 

 

 

when I needed the information 1 2 3 4 5
35 Because no one 1 knew was planning to

attend I 2 3 4 5
36 Becauselpreferioleamonmyoun .................... 1 2 3 .t 5
3‘.” Because 1 was not willing to pay the cost

of the registration 1 3 t 4 5
38 Because my friends did not encourage

my participation 1 2 3 4 S
39 Because I didn‘t find out about the activity

2 3 4 5

far enough in advance to make plans .................. 1

Now we would like to ask some questions for statistical analysis. (Circle the item that
applia. or fill in the requested information.)

1 What is your gender? (Circle one item.)
1 MALE 2 FEMALE

2 in which age category are you? (Circle one item.)

I 18-24 4 45-54
2 25-34 5 55-64
3 3544 6 65&over

3 in what county are most of your agricultural products raised or produced?

COUNTY NAME:

 

4 List the number of acres or square feet that you manage? (Include owned and rented.)

ACRES (FARM. ORCHARD. ETC.)

OR
SQUARE FEET (GREENHOUSES. ETC.)

179

5 What is your highest level of education completed? (Circle one item.)

MLWIQH

\D

ELEMENTARY SCHOOL 6 Z-YIZAR COLLEGE DEGREE

SOME HIGH SCHOOL 7 4-YEAR COLLEGE DEGREE

HIGH SCHOOL DIPLOMA OR GED 8 SOME COLLEGE GRADUATE WORK
SOME COLLEGE 9 GRADUATE DEGREE

TECHNICAL OR

TRADE SCHOOL CERTIFICATION

What is the PRIMARY enterprise on your farm or business? (C irclc only one item )

1 CASH CROPS 5 FRUIT

2 DAIRY 6 VEGETABLES

3 LIVESTOCK ( EXCEPT EQUINE) 7 NURSERY. GREENHOUSE. ETC
4 EQUINE 8 OTHER

 

What is your average annual gross sales from the farm or business? (Optional)

50,000 - 99.999
100,000 - 249,999
250,000 - 499,999
$500,000 OR MORE

LESS THAN $2,500
2.500 - 4.999

5,000 — 9,999

10,000 - 24.999
25.000 - 49,999

khhuto—
NOW*\lC~

Do you consider yourself a part-time or full-time producer? (Circle one item.)
1 PART-TIME PRODUCER 2 FULL-TIME PRODUCER
in addition to your farm or business. do you have an off-farm job? (Circle on: ire».- )

1 NO 2 YES

Finally. please list the item tiiat is most important to you in your decision to participate in an
educational activity. (Write these on the line below.)

ITEM :

 

Feel free to list below. any barriers to your participation in educational activities that were not
mentioned in this questionnaire? Also list any other suggestions you have.

 

 

THANK YOU I!
When you are finished with this questionnaire. please fold along the dotted line.
tape the survey (do not staple). and place in me mail.

180

 

 

 

 

 

 

 

 

 

0007
Fold on the (me I
”H”
No POSTAGE
AR
IF MAILED
IN THE
UNITED STATES
—
BUSINESS REPLY MAlL _
FIRST CLASS MAIL PERMIT NO. 941 EAST LANSING, Ml _
POSTAGE WILL BE PAID BY ADDRESSES —
—
AEE CENTER FOR EVALUATNE STUDIES
_

MICHIGAN STATE UNIVERSITY
409 AGRICULTURE HALL
EAST LANSlNG, MI 483243902

APPENDIX C

FIRST COVER LETTER

E

w at w
& Em Bee-tho
£10 Autumn: Pd!
“ct-pan Sate may
Ezs: Lansing Maegan
48824-1039

517-355-5530

FAX 517353431

702me
Mam-Wm
Mam

mama-1mm.

MICHIGAN STATE
0 N I v E R s l T Y

March 11. 1997

car Agriculmral or Horticultural Producer:

Producers sometimes like to take part in educational activities. They may do this to
learn a new idea related to the farm or business. They may also want to learn about a
non-farm interest such as a hobby. However. sometimes barriers or other reasons
prevent producers from taking part in these educational activities.

We know very little about these barriers or reasons. The only way to find out is to
ask you. Your thoughts and opinions are important to us. We have invited only 15
or 20 producers in each county of Michigan to take part in this study.

We have enclosed a questionnaire. This questionnaire will take about 12 minutes
to complete. You indicate your voluntary agreement to participate by completing
and returning this qucsnonnaire. We will keep your responses confidential. You
will not place your name on the questionnaire. We have placed a code number on
the questionnaire for mailing purposes only. When we receive your completed
quesuonnaire, we will remove your code number and name from the mailing list.
This will prevent us from sending you a second questionnaire.

Michigan State University will use the results of this study in planning educational
programs for producers. Please return your completed questionnaire by Tuesday.
Mamh 13 10:07. When completed. simply fold the questionnaire in half as
instructed. Then tape it shut and drop it in the mail box. We have already paid the

postage.
Thank you very much for taking part in this study.

Sincerely.
_- f/ I/ f o/
t, T Xi ‘ 7 l
\i <57" .

/
Mark F. Hansen
Extension Agricultural Agent

181

APPENDIX D

POSTCARD REMINDER

Dear Agricultural or Horticultural'Producer:

Last week we mailed you a survey entitled “Participation in
Educational Activities”. If you have already completed and
returned the questionnaire, please accept our sincere thanks. If not,
please do so as soon as possible. Your response is very important
to us. The opinions of Michigan producers will be used to help
improve the way that educational activities are offered. If by some
chance you did and receive the questionnaire or it was simply
misplaced, another questionnaire will be mailed to you within the
next two weeks. Thanks for your assistance.

Sinc @fl {1 2
7 / \/ A ‘T-

Mark/F. Hansen
Michigan State University Extension

 

 

182

APPENDIX E

SECOND COVER LETTER

 

worm

5mm
“OWN
”Slum
tam-1m

517-3555!)
FAX 517.3534”

mums-m
wean-mm
mam

talcum

April 1,1997

Dear Agricultural or Horticultural Producer:

Three weeks ago, we mailed you a questionnaire entitled “Participation in
Educational Activities”. We have now heard from a number of producers.
However as of this writing, we have not received your response. If you have
already responded. thank you. We realize that you are very busy, and probably
receive many surveys in the mail. However, very little research has been done to
understand the barriers that farmers face in trying to attend educational aCtivities.
That is what we are attempting to study. We have invited only 15 or 20 producers
in each county of Michigan to take part in this study.

We have enclosed a questionnaire in case the first one was misplaced. This
questionnaire will only take about 12 minutes to complete. You indicate your
voluntary agreement to participate by completing and returning this questionnaire.
We will keep your responses confidential. We have placed a code number on the
questionnaire for mailing purposes only.

Michigan State University and other organizations will be able to use the results
of this study in planning educational programs for producers. Please return your
completed questionnaire by Tuesday. W. When completed, simply
fold the questionnaire in half as instructed. Then tape it shut and drop it in the
mail box. We have already paid the postage.

Thank you very much for taking part in this study.

Sincerely.
,’l

/ /"c’%/J “W

Mark F. Hansen
Michigan State University Extension

 

183

 

APPENDIX F

CORRELATIONS BETWEEN SOURCE FACTOR SCORES
AND SOCIODEMOGRAPHIC VARIABLES FOR
FARMER PARTICIPANTS

Table 36. Correlations Between Deterrent Factors and Sociodemographic
Variables for Farmer Participants

Source Factors

 

 

Socio- l 2 3 4 S 6
Demographic Lackof Lack of Time Low
Variables Course Confl- Con- Personal Scheduling

Relevance deuce straints Priority Constraints Cost

Gender -.014 -.154"‘ -.062 -.123"' .013 .103

Age .077 .127“ -.030 .046 -.041 -.027

Level of

Education -.077 -.282" -. 105 -.091 .012 -.013

Annual

Gross Sales .137“ .018 .176" -.048 -.171" -.087

Part-time or

Full-time .040 -.008 .170" -.063 -.269“ .039

Off-Farm

Job -.052 -.085 -.021 -.064 .301" -.064

*p < .05 '"p < .01

184

APPENDD( G

CORRELATIONS BETWEEN SOURCE FACTOR SCORES
AND SOCIODEMOGRAPHIC VARIABLES FOR
FARMER NON-PARTICIPANTS

 

Table 37. Correlations Between Deterrent Factors and Sociodemographic
Variables for Farmer Non-Participants

 

 

Source Factors
Socio- l 2 3 4 s 6
Variables Course Confi- Cou- Personal Scheduling

Relevance deuce straints Priority Constraints Cost
Gender .115 .146 .174 -.O78 -.O93 .341 "
Age

.039 .229“ -.113 -.061 -.191 -.001
Level of
Education -.065 -.O94 .014 -.238* .056 -.127
Annual
Gross Sales .050 -. 150 .211 -. 187 -.120 .104
Part-time or
Full-time -.065 -.096 .133 -.184 -.214 .106
Off-Farm
Job .068 -.107 .056 .218 .401 ** .098
*p< .05 "p< .01

185

APPENDIX H

OPEN-ENDED RESPONSES TO
THE QUESTIONNAIRE

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE

The following comments are in response to the open-ended question on Page 7 of the
questionnaire, i.e., “Finally, please list the item that is most important to you in your
decision to participate in an educational activity.”

Location, Tinting, Subject.

Widow and unable to do anything difi‘erent.

Having it during the time (hours) I can attend.

Having enough time to go to class.

Location of site as well as time of day.

Time, Location.

Will it help me make more money.

l-Location, 2-Subject Matter, 3-Appropriate for field of endeavor, 4-Timely
Subject.

Need to know or improve.

I really want to hear about some agricultural ideas unrelated to commercial
activities.

Location of educational activity.

Location and time of year.

Item most important for participation.

Benefits it will have for me.

Interest in topics.

Bottom line.

Knowledge to improve farming practices, RUP credits, working cooperation and
rapport with Extension agent.

I don't know about these programs.

Growing information with details, facts, new organic ways.

Ifit interests me and have time to attend; distance.

I'm 86 years old and must conserve extra [?]."'

Ofl‘ers information 1 need.

Time of year, of day & location.

Money & Time.

I feel the college professors are well-paid and the opportunity to come to local
areas should be rewarding.

Who the speaker is; pe0ple that are in industry often more knowledgeable and can

relate better.

Equine.

Do I really need it?
Leader, Location, Cost.

186

 

187

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE

(Continued)
Need to know.
Information on the subject
If the activity is to the point and not dragged out.
Location, costs, subject.

Advance notice & content of meeting to be held.
Convenience, time.

Location.

Needed to continue my business.

Time of activity - it cannot interfere with job.

22” Rows.

Usefulness in improving personal performance.
Who the speaker or teacher will be.

We need to be able to apply the topic to our farm and it needs to be a new idea.
Person giving activity & subject.

Convenient time & place.

Topic.

Timing with chores.

Time of the year - winter months only.
Aninterestinwhatisbeingtaught.

Item most important for participation.

To learn; to talk with other producers.

Time of day of the meetings.

Free time and no conflict with other activities.
Benefit to me.

Something that would benefit me at 64 years old.
Whether or not the subject matter is of primary concern to me at the time.
Time - I don‘t have enough time!

Subject.

It's value to me.

Place - time - educational activity.

Time & Location.

Knowledge (useful) gained vs. time commitment.
It is educational so I learn.

Subject to be covered, where & when.

Time of day & Location (how far away).
Information I needed, had time to participate & handy.
Location and a convenient time (evening).
Expenses, travel, time.

Health permitting.

188

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

[7] - Item # 23 (”Because the available activity did not seem useful or practical")
Quality of program.

To stay abreast & learn of new products & methods.

Item #23 ("Because the available activity did not seem useful or practical. ").
Time of day. Time of year & location.

Topics of current interest relating to profitability of my operation.
Is it beneficial to my family or business?

Farm coverage to allow me to go.

Value of info presented.

Perceived program value to my goals.

To learn something.

Length & Time.

To know more about a product or practice that's taught.
Available time.

Ifit pertains to my farm operation.

Interest in the subject.

Time to attend.

Will it benefit my business?

Acquisition of knowledge.

Information up-dates, exchange ideas.

Time of activity. Most conflict with ofi‘-farm job.

Relate to other producers.

To keep up on new system to use.

My problems with alfalfa. (Bugs, weeds, & disease).

To enhance my knowledge to farm better & more eficient.
Encouragement - Approval - Cost.

Item most important for participation.

Reforestation.

How it afi‘ects my business financially.

Speaker & Material.

Education in my line of production.

Time. Have a part-time job as a [7], not sure when I may have to work for them.
New technology.

Scheduling plus business interests.

Useful on my farm.

Keeping up with the times.

Bee keeping.

Computer knowledge for farming & fun.

Total time invested vs. knowledge/materials gained.

189

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

I just don't have much time.
The time of day planned
Time of year.
Cost vs. what will it do for my farm. Loose $200.00/day in wages & travel if I
attend vegetable convention in Grand Rapids.
Time of year.
To keep abreast of the agricultural changes.
Ifthe classes will be up to level anticipated to be.
Money.
Self Improvement.
Topic.
I look at specific results & the level of information.
Time.
Continue to learn new things.
Time.
Meeting my farming needs.
Factual information that is up to date.
Personal interest in topic and amount of creativity involved in participation. ..don't
like lecture style...
Topic and content.
Time and purpose.
Day of week & time of day.
Soil quality/health - will it help to reduce ofi‘-farm inputs - less pesticide/fertilizer
purchases.
Time available.
Location, Interesting, Timing, Subject.
Keeping my mind open to new ideas.
Had time. -
Farming.
Item most important for participation.
I'm a hobby farmer just to have something on the farm.
Timing and length of season.
Information to be given (topic).
Specific interest and location.
Cost, topic, of meeting, time of year or day.
Knowing when & where the activity is.
Cost, Location, Time.
Review & New Technology.
Location.

190

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

To learn more about my line of work.

Of importance to farm operation.

The content of activity.

Convenience.

Has to be valuable to my occupation.

Is interesting to me.

Time & Location.

Someone who has experience.

Interested in topic.

Will it do me any good?

Computer up-dates are needed in agriculture, fast if not yesterday.
Topic & Time.

Location.

Ifl have the time and it is an interesting subject, I will attend.
Knowledge & Special Activities.

Convenient Location.

Ifit is applicable.

Information for crops I grow; close to home within 30-40 mile drive.
Interest in the subject.

Location and the time of day - prefer evenings.

Specific subject matter.

Covet crop information is real helpful.

Interest in subject matter and opportunity to learn.

Day of week & time of day.

Location.

Amt of practical-hands on experience the instructor has - not # of degrees or # of
papers he has published.

The program relevance.

Item #3 (”Because there was no one at the farm or business who could cover for
me. ”).

Location & Length.

[H can use what I learn.

Time.

Location.

Keep informed about new technology in the greenhouse business.
Item most important for participation.

I like to see new things and learn new [languages]."

To obtain license.

Being able to work into my schedule.

 

191

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

Usefirlness.

Location.

I don't have time for many outside activities.

Money.

Am I going to learn anything new.

To learn more.

Timing - medication at time needed.

Something related to improving 8 return in our farming activity - to do a better
job.

Ifit pertains to the crops I raise.

Retired.

No—till.

We are retired (but eamings from farm).

Value of material to be covered.

Relevancy to our business.

Business, the more I learn the less I have to sub—out labor.
Location.

Whether I have a sitter for my children.

Time, location & cost.

Ifit covers information that I can use.

Time available when not working at full-time job.

Learn more about something.

Has to do with the crops I grow.

Ifit is interesting and I can learn about something new.

Cost and if it is interesting.

Finding the time in my schedule to attend.

Presentation topic & content.

Time & Location.

Topic must include finit & vegetable crops which I propagate.
Time of day - prefer morning.

Time of day.

Will it be interesting and be an addition to what I already know.
If I want to do it or not.

In Florida 8 months, 4 months in Michigan.

Notification of educational activity, time to plan for it.
Subject, time, location, in that order.

I know little or nothing of your courses. No advanced notice of courses.
To learn more on subjects that are hard to get information on.
How to be most productive-dairy cows or steers - or get out.

fi.‘

192

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

I wanted a session on fruit-you had one 10 years ago. Everything I learn extra 1
pick up from Pro-Farmer.

Increase my knowledge in that topic.

Item most important for participation.

Time of day.

Time.

Evening classes.

Get the pesticide license.

Topic & when oflered.

Cost.

Computer Training.

Item #4 (Because the farm or business comes first and I didn‘t think I should be
gone).

Topic & time available.

Relevant and specific - most activities seem to general.
Starting and ending time, as I have cattle to feed.

Less than 25 miles fiom home between 10:00 am and 4:00 pm.
Subject content.

Immediate need for information ofi‘ered.

Ifl have the time to go.

Location close to home - within one hour.

Topics.

Location - how helpful it will be to my situation.

Didn't know about any. ‘

Cross breeding beefcattle & raising hay.

Need for knowledge.

I am going to be 70 years old this year; I have enough interest without firrthering
my education.

Usefulness and timing.

Solution to control of honey bee mite problems.

Being more profitable.

Will I benefit?

Ifit is scheduled during evening hours.

Topic, Location.

Value of information presented.

Becoming a better servant of our Lord Jesus Christ.
Something that fits my needs.

Anything to improve general farming.

Personal interest and potential for learning something new.

193

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

Money.

New concepts and products.

Directly benefits the business.

Problem solving - Continue to involve MSU stafi‘ members.
Item most important for participation.

Greenhouse.

Information (useful), deregulation.

Lack of information on when [activity is held].

Curiosity.

Whatever activity involves to learn more about my business.
Forestry.

Free time to attend.

Interest.

My need to know.

Something that pertains to my dairy.

None.

People 1 meet sometimes give me more information than I receive at the meeting.
None.

IfI feel it will help my business.

Useful [information].

Information that is practical - that I can use.

Direct interest in our operation - and quality instructors.
Location.

None.

Time of day.

Time, place, subject.

Need evening meetings.

I just don‘t have the time.

To be mailed info on time.

Something new worthwhile learning.

Hearing problem.

Curiosity; desire to learn more.

Length of time.

Money.

No activities are of interest to me.

Need information to make a decision.

Dairy management.

Learn how to make a profit.

Reasonable cost/critical information/as little time as possible.

194

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE

(Continued)
Application.
Interest.
Do I feel I will learn something at the meeting.
Needed information.

Item most important for participation.

Personal interest in the subject being covered.

Improved production.

Will it help me be a better person in general.

The usefulness of the information.

Location and time of year.

Items #7, #20, and #30.

Ifit helps increase cash flow.

Future plans for feeding the world.

Computer - knowledge for record keeping.

Time and distance away.

Item #8 (. activity that was offered did not seem interesting enough).
Time of year and day.

Type of class.

Better way of farming.
IleamnewthingsallthetimeaboutthingslalreadythoughtIknew.
Convenience and content.

Information.

Being able to take ofi‘ work at second job to attend.

If it is something I am interested in and I can put the information to use.
Applicability/relevance. I've heard too many of "same guy - same [explicative]”
presentations, [e.g., Kg/ha].

Time and appropriate topic.

Ifit is something I can use.

Just wanting to learn something difi‘erent.

Those that solve problems.

Getting too old.

Ag laws.

Relevance to my operation.

Interesting.

Had to work out [03 the farm] all my life to keep the farm.

Work horses.

None.

Needs to be practical in everyday life - Item 23.

To learn in order to do a better job.

19S

OPEN-ENDED RESPONSES TO THE QUESTIONNAIRE
(Continued)

Knowledge is the only thing that can't be stolen fiom you.

Item most important for participation.

Pesticide control.

I'm not interested in farm related educational activities.

Interest - new ideas.

Must not be too general.

The need for information offered.

Some informational activity and good speaker.

Whether or not the topic is useful to my farm production.

Need in-depth information on specific topics.

Relevant information to our life.

Knowing about them.

Type; time of year should be winter.

If it would be held on weekends, not weekdays.

Have educational dairy or agricultural meetings closer to home.
Available time.

Learn something that will be useful.

Second request.

Time.

Time.

Something that will benefit me and my family.

Lack of time.

"Ignorance” - the older I get, the more I realize that I don‘t know. I would
attend. . .if more were ofl‘ered.

[Would like to] Learn how to write.

Location and time.

Fore-knowledge of the event & enough information showing the activity
worthwhile.

Conflict with my ofl‘-farm job.

Farming [but] retired.

I want to be knowledgeable in this area [dairy farming] as much as possible.
Location (travel time).

Educational activities scheduled during workday.

Learning - especially if it afl‘ects the bottom line or the environment that I live in.

"' [7] - Unable to read handwriting and unsure of word or phrase.
"' "' [ ] - Items in brackets were added or rephrased to clarify comment.

BIBLIOGRAPHY

BIBLIOGRAPHY

Ary, Donald, Lucy Cheser Jacobs, and Asghar Razavieh, 1990, W
infiltration, Fourth Edition, Holt, Rinehart and Winston, Inc, Chicago, Illinois,

536 pp.

Minn, George H» 1984, WWW Food and
Agriculture Organization of the United Nations, Rome, 148 Pp.

Babbie, Earl, 1983, WW1), Third Edition, Wadsworth Publishing
Company, Belmont, California, 551 Pp.

Bemsten, Richard H., 1993, Changes in Michigan Agriculture: 1950 to 1987, Status and
Potential of Michigan Agriculture - Phase I, Special Report 34, Michigan
Agricultural Experiment Station, Michigan State University, East Lansing,
Michigan, 31 pp.

Brewer, Frank L., 1996, Graduate Class: International Agricultural and Extension
Systems, Agricultural and Extension Education, Michigan State University, East
Lansing, Michigan.

Brewer, Frank L., and Mark F. Hansen, 1989, Ag Interface Meeting Series, Extension
Agriculture and Marketing Programs, Michigan State University Cooperative
Extension Service, Michigan State University, East Lansing, Michigan.

Brown, Ronald PI, Paul A. London, and Harry A. Scarr, 1992, 1992 Census of
Agriculture, Michigan State and County Data, Part 22, Volume 1 Geographic Area
Series, Bureau of the Census, Economics and Statistics Administration, US
Department of Commerce, 509 pp.

Carey, Donald, 1992, Review of Advisory Services, Report of the Review Group
Established by the Director of Teagasc, Teagasc, Dublin, Ireland, 12 Pp.

196

 

197

Cooney, Graham, 1987, Private Farm Management Advisory Services, Farm Consultancy
Services in the Free Market Economy, Special Publication - Tussock Grasslands
and Mountain Lands Institute, University of Canterbury, Canterbury, New
Zealand, Pp. 21-25.

Cross, K. Patricia, 1981, Adults As Learners: Increasing Participation and Facilitating
Learning, Jossey-Bass Inc. Publishers, San Francisco, California, Pp. 300.

Curley, Thomas and Lorcan O'Toole, 1993, Personal Interview on the Aspect of
Extension Fee-For-Service Programs via Teagasc, Dublin, Republic of Ireland.

Darkenwald, Gordon 6., and Sharan B. Merriam, 1982, MW
Emilia, Harper & Row, Publishers, New York, 260 Pp.

Darkenwald, Gordon 6., and Thomas Valentine, 1985, Factor Structure of Deterrents to
Public Participation in Adult Education, Adult Education Quarterly, Vol. 35, No.
4, Summer 1985, Pp. 177-193.

Davis, Lansing J ., 1988, Deterrents to Bank Managers’ Participation in Continuing
Education, Doctoral Dissertation, Rutgers, The State University of New Jersey,
New Brunswick, New Jersey, 132 pp.

Dillmarr,DonA., 1978, u '

; _ - . r,-
erey & Sons, New York, 325 Pp.

‘5'" U -t l 0", 10h“

   

Dunteman, George H., 1989, Principal Coponents Analysis, Sage Publications, Inc.,
Newberry Park, California, 96 pp.

Elliot, Jack, 1990, Graduate Class in Research Design, Fall Quarter 1990, Department of
Agricultural and Extension Education, Michigan State University, East Lansing,
Michigan.

Elliot, Jack, 1991, Graduate Class in Research Instrument Development, Winter Quarter
1991, Department of Agricultural and Extension Education, Michigan State
University, East Lansing, Michigan.

Essman, Elizabeth Rossi, 1994, Deterrents to Participation in Continuing Education
Programs Among University-Based Nurse Faculty, Doctoral Dissertation,
Syracuse University, New York, 241 pp.

Ferris, John N., 1994, Some Salient Facts fi'om the ”1992 Census of Agriculture",
Department of Agricultural Economics, Michigan State University Extension,
Michigan State University, East Lansing, Michigan, Memo dated July 27, 1994.

 

198

Ferris, Maxine S., 1992, Agricultural Technology Trends and Outlook (Briefs): Trends in
Education and Information Dissemination, Special Report 35: The Status of
Michigan Agriculture and Its Resource Base, Status and Potential of Michigan
Agriculture - Phase 1, Michigan Agricultural Experiment Station, Michigan State
University, East Lansing, Michigan, Pp. 63—64.

Ferris, Theodore A, 1994, Letter Response to DACUM, Department of Animal Science,
Michigan State University, East Lansing, Michigan.

Finley, Eddy and Robert R Price, 1994, MW, Delmar Publishers,
Inc, Albany, New York, 344 pp.

Forbes, Allen, 1934, WWW Printed for the State Street
Trust Company, Boston, Massachusetts, 94 pp.

Fraenkel, Jack R, and Noman E. Wallen, 1993, MW
Education, McGraw-Hill, New York, 571 pp.

Giacoma,Pete,1989, I..-

,Neal-Shurnan Publicshers, Inc, New York, 200
PP-

 

Gould, MD, and G. Cooney, 1987, MAFTech, Farm Consultancy Services in the Free
Market Economy, Special Publication - Tussock Grasslands and Mountain Lands

Institute, University of Canterbury, Canterbury, New Zealand, Pp. 9-19.

Guertin, Wilson H., and John P. Bailey, Jr., 1970, W
Analysis, Edwards Brothers, Inc, Ann Arbor, Michigan, 472 pp.

Harman, Harry H., 1960, W, The University of Chicago Press,
University of Chicago, Chicago, Illinois, 469 pp.

Hatter, Don, and Glen Hass, 1992, Commercialization of British Extension System:
Promise or Primrose Path, Paper Presented at the Canadian Society of Extension
Conferrence in Brandon, Manitoba, Journal of Extension Systems, Vol. 8, No. 1 &
2, Pp. 37-44.

Healy, Thomas and Associates, Ltd., 1991, National Cattle Farmers Survey: Farmers'
Attitudes to TEAGASC Advisory Services, Research Report Prepared for
Teagasc, Dublin, Ireland, 33 Pp.

Heinze, Kirk L., 1996, Personal Conversation Regarding Appropriate Guidance to Focus
Groups, Department of Agricultural and Extension Education, Michigan State
University, East Lansing, Michigan.

199

Hepp, Ralph E., 1992, Farm Financial Management Workshop, Kellogg Foundation
Grant, Department of Agricultural Economics, Michigan State University, East
Lansing, Michigan.

Hobbs, Daryl, 1992, The Rural Context for Education: Adjusting the Images, Education

WW Krieger Publishing
Company, Malabar, Florida, Pp. 21-41.

Houle, C. O., 1961, Ihflnmridnghfind, University of Wisconsin Press, Madison,
Wisconsin, 87 pp.

Hubbard, Charles L., and Clark A. Hawldns, 1969, Theory of Valuation, International
Textbook Company, Scranton, New Jersey, 249 pp.

Imig, Gail L., 1995, Satellite Teleconference on Area Of Expertise, Michigan State
University Extension, Michigan State University, East Lansing, Michigan.

Johnstone, John WC, and Ramon J. Rivera, 1965, MW Aldine
Publishing Company, Chicago, 624 pp.

Kenyon, Gene. Study Strata. [Online] Available email: hansenm@msue.msu.edu fi'om
gene_kenyon__at_nass-mi@nass.usda. gov, October 20, 1998.

Kirley, Thomas, 1994, Implementation of the Reorganised Advisory Service - National
Overview, Notes of a Meeting of Chief Agricultural Oficers, Teagasc, Dublin,
Ireland, 6 Pp.

Kirley, Thomas, and Lorcan O'Toole, 1993, Informational Interview on Fee-For-Service
Programs in Teagasc, Dublin, Ireland.

Krueger, David, and Murari Suvedi, 1996, Program Evaluation Workshop, Department of
Agricultural and Extension Education, Michigan State University, East Lansing,
Michigan.

Krueger,RichardA, 1994, .~ '2 '
Publications, Thousand Oaks, California, 255 pp.

 

Leholm, Arlen G., 1995a, Fall Extension Conference - October 1995, Michigan State
University Extension, Michigan State University, East Lansing, Michigan.

Leholm, Arlen G., 1995b, Michigan State University Extension Budget Information,
Extension Agriculture and Natural Resources Program Area Stafl‘ Meeting,
Michigan State University, East Lansing, Michigan.

200

Levine, S. Joseph, 1992, Graduate Class: Planning Educational Programs, Department of
Agricultural and Extension Education, Michigan State University, East Lansing,
Michigan.

Lillis, Seamus, 1993, Personal Conversations on Fee-For-Service Programs in Teagasc,
Dublin, Ireland, and East Lansing, Michigan.

Martindale, Cameron J. and James B. Drake, 1989, Factor Structure of Deterrents to
Participation in Ofl‘-Duty Adult Education Programs, Adult Education Quarterly,
Vol.39, No.2, Winter 1989, Pp. 63-75.

Maslow, Abraham H., W, Second Edition, Harper Row,
Publishers, New York, 369 pp.

Merriam, Sharan B. and Rosemary S. Cafl‘erella, 1991mm
Cnmpnehensiyefiride, Jossey-Bass Publishers, San Francisco, California, 376 Pp.

Miller, Larry E., and Keith L. Smith, 1983, Handling Nonresponse Issues, Journal of
Extension, September-October, 1983, Pp. 45-50.

Mitchell, Robert Cameron, and Richard T. Carson, 1989, W

W Resources for the Future, Washington,
D. c., 463 Pp.

NON-'48, M31133 1, 1991. MW SPSS. 1110.,
Chicago, Illinois, 499 pp.

Norusis, Mattie 1., 1992, SEW; SSPS. Inc,
Chicago, Illinois, 580 pp.

Pscodna, Stephen, 1996, Personal Telephone Correspondence to the Michigan
Agricultural Statistics Service on Generalizing Data fiom MASS Mailing Lists,
Michigan Department of Agriculture, Lansing, Michigan.

Rogers, Everett M., 1983, W 3rd Edition, Free Press, New York,
453 pp.

Scanlan, Craig L., 1982, Factors Deterring Allied Health Professionals in Continuing
Education, Doctoral Dissertation, Rutgers, The State University of New Jersey,
New Brunswick, New Jersey, 533 pp.

Scanlan, Craig S., and Gordon G. Darkenwald, 1984, Identifying Deterrents to
Participation in Continuing Education, Adult Education Quarterly, Vol. 34, No. 3,
Spring 1984, Pp. 155-166.

 

201

Shapley, Allen, 1992, Agricultural Technology Trends and Outlook (Briefs): Farm Labor,
Special Report 35: The Status of Michigan Agriculture and Its Resource Base,
Status and Potential of Michigan Agriculture - Phase 1, Michigan Agricultural
Experiment Station, Michigan State University, East Lansing, Michigan, Pp. 63-
64.

Scott, RoyV., 1970, Ir °- - -
University of Illinois Press, Chicago, 362 pp.

 

Tough, Allen, 1971, It' ‘ vii ' ’ '
W Research in Education Series No.1, The Ontario
Institute for Studies in Education, Toronto, Ontario, 190 pp.

 

Vergot, Pete III, 1991, Southwest Michigan Fruit and Vegetable Farmers' Use of Sources
and Channels to Gain Information in Regard to New Agricultural Practices,
Doctoral Dissertation, Department of Agricultural and Extension Education,
Michigan State University, East Lansing, Michigan, 149 pp.

Vines, C. Austin, ande A. Anderson, 1976, W
Cmnmitmentjefieeple, Journal of Extension, Madison, Wisconsin, 236 pp.

Weimer, David L, and Aidan R. Vining. 1992, W
2nd Edition, Prentice Hall, Englewood Clifl‘s, New Jersey, 424 pp.

Weischadle, Mary Ann Piscopo, 1988, Effects of Attitudes and Deterrents on Participation
in Continuing Education by Real Estate Professionals, Doctoral Dissertation,
Rutgers, The State University of New Jersey, New Brunswick, New Jersey, 114

pp.

Westermarck, Harri, 1987, Advisory Consultancy in Finland: Farmers Attitudes Toward
the Charging Systems for Extension Services, Journal of Extension Systems, Vol.
3, Pp. 3 5-41.

 

"‘uhtutu“