ERIC ED369921: Factors Related to the Motivation of Extension Agents in Kenya's Rift Valley Province. Summary of Research 76.

13 Pages·1994·0.52 MB·English

by  ERIC

DOCUMENT RESUME
CE 066 334
ED 369 921
Mwangi, John G.; McCaslin, N.
L.
AUTHOR
Factors Related to the Motivation of Extension Agents
TITLE
in Kenya's Rift Valley Province. Summary of Research
76.
Ohio State Univ., Columbus. Dept. of Agricultural
INSTITUTION
Education.
PUB DATE
94
NOTE
13p.
Information Analyses (070)
PUB TYPE
MFOI/PC01 Plus Postage.
EDRS PRICE
Adult Education; *Extension Agents; Extension
DESCRIPTORS
Education; Foreign Countries; *Job Satisfaction;
*Motivation; Quality of Working Life
*Kenya
IDENTIFIERS
ABSTRACT
Job satisfaction factors related to motivation of a
random sample of 325 agents in Kenya's extension service in the Rift
Valley Province were identified. Data were collected using a
group-administered, closed-ended, forced-choice questionnaire with
105 job satisfaction, 10 motivational, and 10 personal
characteristics items between January and March 1993. Agents were
also interviewed as a group in each district. Findings indicated
motivation
personal characteristics were not as important for agents'
Eight job satisfaction factors
as were the job satisfaction factors.
decreasing order of
were important for agents' job satisfaction. In
importance, they were as follows: evaluation, dependable supervisors,
work incentives, pay, praiEe and work location, housing and
transportation, job security, and administration and supervision.
Five of the eight job satisfacti,on factors were important for agents'
motivation. In decreasing order of importance, they were as follows:
dependable supervisors, pay, job security, evaluation, and
administration and supervision. One cause of agents' frustration and
ignored
low job motivation was their perception that merit was being
in selecting candidates for inservice staff training. Job
satisfaction and motivation were related but different.
Recommendations for extension managers and suggestions for future
research were made. (Contains 38 references.) (YLB)
***********************************************************************
Reproductions supplied by EDRS are the best that can be made
from the original document.
*
***********************************************************************
SR 76
U.S DEPARTMENT OF EDUCATION
111fice or tdocalional Resean.n and imceovement
Summary of
I O :ATIC)NAL RESOURCES INFORMATION
% E N FE IR IERICI
PERMISSION TO REPRODUCE THIS
thiS document nas been reDrOduCed as
Orpanotatlian
rt.( rived lino, the perSon or
esearch
MATERIAL HAS BEEN GRANTED BY
.niginaling it
Minot r handes have been ma0e lo improve
I
.eDrOduction quality
Ihisii0Cu
e Po.nts of toe. 01 opinions slated in
.nnnt rio 001 necessarily meolesent Otbcial
pOlICY
oOsttoun
OE
TO THE EDUCATIONAL RESOURCES
INFORMATION CENTER (ERIC)"
Department of Agricultural Education
The Ohio State University, Columbus, Ohio 43210-1067
FACTORS RELATED TO THE MOTIVATION OF EXTENSION AGENTS IN
KENYA'S RIFT VALLEY PROVINCE
John G. Mwangi and N. L. McCaslin
it
motivational appeal but, if threatened later,
Many scholars believe that motivation is
regains potency and remains dominant until it is
mainly responsible for differential staff perfor-
Maslow advised managers to provide
satisfied.
mance, that it changes as time and
conditions
different rewards and counselling to motivate
change, is dependent on incentives that the staff
different workers. He indicated that a fulfilled need
value and believed to be attainable with increased
does not motivate, but other researchers (Glassman,
individual performance, and that it is high when
1978; Heneman et al., 1980; Buford & Bedeian,
staff frustration is minimal (Moris, 1987; Grossnickle
1988; Kreitner, 1989; Davies et al., 1990) have
& Thiel, 1988; Beder, 1990; Cohen, 1990; Schmincke,
shown that satisfying self-actualization needs in-
1990; Winslow, 1990; Watanabe, 1991). Managers
creases motivation.
need to know what motivation is and how they can
from
use it to obtain the desired work performance
Unlike Maslow, Herzberg (1959 & 1972) siressed
Lawler III
their subordinates (Kreitner, 1989).
the need for a favorable work environment saying
(1973) indicated that if little ability is required and
that enriched jobs rather than pay, supervision, and
people have the same ability, skills, and training,
other environmental factors were the key to moti-
positive staff motivation is the singl e most impor-
vation and job satisfaction. Herzberg believed that
tant determinant of effective job performance.
challenging, enriched jobs motivated employees
Besides being more productive, mr,tivated workers
more than dull, routine jobs. He advised managers
equal or excel any standards set by their superiors
to redesign jobs to provide opportuniiies for indi-
(Dowling & Sayles, 1971).
vidual achievement, recognition, responsibility,
advancement and personal growth.
However,
Motivation is the psychological process that
research (Buford & Bedeian, 1988; Kreitner, 1989)
gives purpose, direction, and intensity to behavior
has shown that professional employees have differ-
In this study, motivation is
(Kreitner, 1989).
ent work preferences for which Herzberg did not
assumed to be the reason individuals desire to excel
account.
in their work.
An individual directs personal behavior to-
Maslow (1943) suggested that people are
ward pleasurable and away from undesirable
motivated by five categories of needs. Starting from
Vroom suggested that
outcomes (Vroom, 1964).
the most basic and first to be satisfied, these needs
motivation was determined by perceived probabili-
safety needs,
are physiological or survival needs,
ties of success, and increases as an individual's
social or love needs, self-esteem or ego needs, and
perceived effort-performance and performance-re-
Maslow indicated that
self-actualization needs.
If employees believe
ward probability increases.
higher-level needs emerge after lower-level needs
their actions would lead to rewards that they value
have
are satisfied and that although a person can
and have a good chance of getting them, they will
several needs at once, only one need can dominate
To be effective, he suggested, all
A satisfied need loses its
work harder.
at any one time.
2
2
Second, current information on factors
1991).
rewards must be linked to performance, goals must
related to agents' motivation in Rift Valley Province
be reasonable, and the outcomes must be negoti-
was lacking and it was in Kenya's public interest
ated. He advised managers to identify, support and
to promote agricultural production and efficiency
reinforce individual perception by linking appraisal
through Extension (Kenya Government, 1986 &
to professional and personal development.
Furthermore, as times and conditions
1990).
change, past motivational strategies become inef-
Adams and Rosenbaum (1962) indicated that
Consequently,
fective (Buford & Bedeian, 1988).
treating employees inequitably lowers their motiva-
identifying agents' needs regularly helps provide
tion and performance. They advised managers to
meaningful, motivational, staff incentives. Third,
make inputs required for outcomes as explicit as
modern management practices are shifting manag-
possible for staff to perceive them as equitable.
ers' emphasis from control to concern for people's
creativity, interests and welfare (Steiner, 1973;
Skinner (1969) reported that behaviors result-
Caron, 1989; Rouche et al., 1989; Yukl, 1989; Perry
ing in desirable consequences are likely to recur
& Wise, 1990; Winslow, 1990; Watanabe, 1991).
while those that result in undesirable consequences
Lastly, inadequate staff incentives make it harder
What an organization
are less likely to recur.
for many Extension systems, particularly in Africa,
appears to reward, he pointed out, is the behavior
to fulfill their responsibilities (Moris, 1987).
that will be seen as the model for success. Because
what is rewarded will be repeated, Skinner advised
Purpose and Objectives
rewarded
managers to state which behaviors will be
and which ones will not, and to tie rewards to
Researchers such as
individual performance.
The primary purpose of this study was to
Kreitner (1989) and Winslow (1990) have indicated
identify job satisfaction factors related to motiva-
that Skinner overemphasized the importance of
tion. Specifically, the study sought to: (1) identify
external outcomes such as pay and promotion,
the agents' personal characteristics; (2) identify
ignored the role of internal outcomes such as
underlying factors of job satisfaction and their
feelings of accomplishment and recognition, and
relative importance; (3) determine the agents'
failed to consider the importance of individual
motivational level; and (4) examine relationships
needs, expectations and values.
between the independent variables and the depen-
On the basis of the study, the
dent variable.
Problem Statement
researcher suggested and recommended ways to
improve agents' motivation and job satisfaction in
Kenya's Rift Valley Province, within the available
Because employees work harder and perform
financial, material, and personnel resources and
better if motivated and satisfied with their jobs
constraints.
1990; Cohen, 1990; Schmincke,
1990;
(Beder,
Winslow, 1990; Watanabe, 1991), Extension man-
Limitations and Basic
their staff to
agers need to know what motivates
(Glassman, 1978;
manage them more effectively
Assumptions of the Study
Glueck, 1978). Preventing motivational problems
before they begin to lower staff performance and
The factors studied were limited to those
disrupt workers' productivity, minimizes employ-
identified by the researcher from the literature and
ees' frustration and boosts their working morale
from personal experience with Kenya's Extension
(Grossnickle & Thiel, 1988). In Kenya's Rift Valley
service. Because only agents who were currently
Province, Extension managers lacked information
working in the Rift Valley Province participated in
and motivation
on factors related to job satisfaction
the study, generalizations were confined to that
which they needed to understand why some agents
group.
performed better than others with similar qualifi-
cations, experience, and abilities. This study was
The researchers assumed that the agents
designed to help provide that information.
understood the questions and responded frankly.
He also assumed that they realized how important
Significance of the Study
their contributions were in helping Extension
staff motivational pro-
managers develop effective
share their actual,
This study was significant for several reasons.
grams, and were willing to
First, public officials need reliable information for
personal feelings.
accountability
accurate decision making and
(Altschuld & Thomas, 1991; Green & McClintock,
3
Methodology
Conceptual Framework
This was a descriptive, correlational research
The framework (see Figure 1) used in this
study. A one-shot case study design (Campbell &
study to identify factors related to agents' motiva-
Stanley, 1963) was chosen to collect data using a
tion was based on two content and three process
group administered, closed-ended, forced-choice
motivation theories. While the content motivation
questionnaire with 105 job satisfaction, 10 motiva-
theories of Maslow and Herzberg deal with what
tional, and 10 personal characteristics items. Prior
motivates people to act in a certain way, the process
to administration, the questionnaire, developed by
motivation theories of Vroom, Adams and Skinner
the researcher, was reviewed and its contents found
are concerned with how people are motivated. The
to be valid by nine faculty and two graduate
framework was developed by reviewing the five
students selected for their background and experi-
motivation theories, combing similar ideas ex-
ence in Agricultural education. A field test indi-
pressed in each, and then summarizing them. For
cated that the questionnaire had face validity.
example, the literature indicated that Maslow's
Agents' job satisfaction factors and personal char-
higher-level needs were ecuivalent to Herzberg's
acteristics formed independent and extraneous
motivators and his lower-level needs were similar
variables respectively. Motivation was the -depen-
to Herzberg's hygienes (Buford & Bedeian, 1988).
dent variable. Of the 2,087 agents who formed the
Based on the literature and personal experience,
frame and accessible population, a random sample
the researchers grouped the factors thought to be
of 325 agents, drawn after stratifying by rank (i.e.,
related to motivation into those that operate within
Agricultural Assistants, Assistant Agricultural
the individual's control (internal factors) and those
Officers and Agricultural Officers) and gender,
outside a person's control (external factors). Per-
completed the questionnaire with a response rate
sonal characteristics (agents' gender, age, marital
A follow-up of agents who failed to
of 84.9%.
status, formal education and years of service) were
complete the questionnaire as scheduled raised the
also studied to determine the extent to which they
response rate to 100%. The questionnaire had an
were related to agents' motivation.
JOB
SATISFACTION
nrrzzruz, recrobts
Achievement
2 Status
3 Hesgonsibility
PENSONAL
4 Work ItStli
CILARACTILIUSTICS
s W.0401000
Fear of Punishment
7 Personal Cresevill
Gender
JOB
Ai*
MOTIVATION
INTERNAL FRCTORS
1
Work 1.44auon
Manua Status
2 Pay
1
3 Transporssuon
Formsl Educatson
a Housing
(level)
5 Working Conditions
Job Sammy
Experience
I
7 Dependable Supernsors
OfEce fatilitsu
(Years of service)
9 Extension Policy
10 Adoomotto0on
11 Suprocion
12 Staff Appreinl
13 Praise
t Personalised Feedback
DEPENDENT
INDEPENDENT
MMUS
EXTRIINEOUS
MIDDLE
INDIES
Una
Conceptual framework for identifying factors related to the motivation
Figure 1.
of Extension agents in Kenya's Mt Valley Province.
4
4
their relative importance for the agents in the Rift
eleventh-grade reading level and a reliability of .79
To determine the underlying
Valley Province.
and .81 for the job satisfaction and motivation
dimensions or factors of the complex phenomena,
items, respectively. Data were collected between
which in this case was job satisfaction, factor
January and March 1993 and were analyzed at The
analysis was used as recommended by Gorsuch
Ohio State University, using the SPSS statistical
package after setting alpha a priori at .05 level.
(1974); Nunnery (1978); Arrindell and van der Ende
Exploratory factor analysis was used to identify
(1985); Ford, MacCallum, and Tait (1986); Norusis
factors related to agents' job satisfaction and
(1990); and Hair, Anderson, Tatham and Black
(1992). A maximum likelihood (common factors)
Following the administration of the
motivation.
factor analysis was conducted since the researchers
questionnaire, the agents were interviewed as a
assumed that the variance of each measured
They were asked five
group in each district.
variable could be decomposed into common and
questions as follows: Were the questions clear? Was
This approach is appropriate
unique portions.
Did the questions
the English understandable?
when the measured variances are assumed to be a .
cover things that interest Extension staff?
Are
linear function of the measured (latent) variables
there things that ought to have been covered that
(Ford et aL, 1986). The maximum likelihood factor
were omitted? Would you say that Extension agents
analysis was deemed appropriate since the analysis
are highly motivated, motivated or not motivated?
was done on a sample rather than a population
Findings
(Norusis, 1990).
The number of subjects recommended for
The findings are related to the four objectives
conducting factor analysis varied from five to ten
of the study. They are presented in sequential order
per item (Gorsuch, 1974; Nrrmary, 1978; and
as follows:
Arrindell & van der End, 1985). Hair et aL (1992)
indicated that factor analysis needed at least a
Objective One
sample of 50 but preferably 100 observations. As
a general rule, they recommended a conservative
Objective one sought to identify the personal
figure of 4 or 5 observations per variable but pointed
characteristics of the agents which included gsnder,
out that in many instances, researchers are forced
age, marital status, formal education and years of
to factor-analyze a set of variables when only a 2:1
Nearly one in four agents was a female.
service.
ratio of observations to variables is available. Since
The agents' mean age was 34.6 years with the
no studies had been conducted on the job satisfac-
youngest and oldest agent being 24 and 55 years,
tion of Kenya's Extension agents, this exploratory
respectively. About 85% of the agento were married,
research used approximately three subjects per item.
14.5% wc...re single, 0.3% were divorced and 0.3%
were widowed.
Decision on the number of factors to extract
before the unique variance begins to dominate
The agents' loWest qualification was a post-
of the
common variance was based on a combination
secondary agricultural 6ertificate while the highest
Latent Root Criterion (eigen value >1) and the
had
was a master's degree. Agricultural Assistants
Scree Test Criterion, as recommended by Hair et al.
served longest, on the average (10.5 years), followed
Using this procedure, eight factors, ac-
(1992).
by Assistant Agricultural Officers (8.5, years) ancL
counting for 24% of the variance in job satisfaction,
Agricultural Officers (5.2 years). The agents' total
were extracted.
33. Nearly 87%
years in service ranged from 1 to
of the agents had worked from 1 to 15 years while
Table 1 indicates the items with their factor
all Agricultural Officers had 10 years or less of
According to Hair et al. (1992) factor
service. Agricultural Assistants had served lor gest
loadings.
loadings greater, in absolute value, than .30, .40
in their current positions (5.5 years). Agents in the
and .50 are significant, more significant, and very
other two ranks had, on the average, served for the
their current
significant respectively. Ford et al. (1986) indicated
same length of time (3.9 years) in
that, as a rule, only variables with loadings greater
positions. About 41% of the agents had not received
than .40 should be considered significant and used
promoted once, 6.5
any promotion, 50.9% had been
In this study, only variables
in defining a factor.
twice, 1.0% three times and 0.3% four times.
with factor loadings of .40 or higher were reported.
Because the factors were correlated with one
Objective Two
another (see Table 2), the Oblimin rotation with
maximum likelihood was used for extraction and to
Objective two sought to identify underlying
arrive at the factor matrix loadings.
factors of job satisfaction related to motivation and
5
Table 1
ROTATED FACTOR MATRIX LOADINGS ORDER OF 35 JOB SATISFACTION ITEMS
ON OBLIQUE FACTORS (N=325)
Factor Loadings
Factors'
Item
8
6
7
5
4
3
2
1
.50
60
.50
87
.49
33
.
.47
45
92
.46
61
.41
.41
20
116
.41
.64
43
.64
28
.50
15
.49
30
.49
57
.44
97
.42
16
-.46
88
.44
37
-.57
24
-.54
84
.45
58
.50
46
.47
36
.47
32
.44
59
.44
52
.76
80
.68
82
.60
51
.60
6
.45
76
.42
99
-.55
22
-.41
26
.47
100
.41
96
3 = Work Incentives, 4 = Pay,
Factor names: 1= Evaluation, 2 = Dependable Supervisor,
Transportation, 7 = Job Security,
5 = Praise & Work Location, 6 = Housing &
8 = Administration and Supervision.
Table 2
UNDERLYING JOB SATISFACTION OF THE
INTERFACTOR CORRELATIONS FOR THE OBLIQUE ROTATED FACTORS
EXTENSION AGENTS IN KENYA'S RIFT VALLEY PROVINCE (N=325)
8
6
7
4
5
3
Factors
2
1
1.00
1
1.00
.13
2
1.00
.12
.10
3
1.00
-.23
-.18
-.01
4
1.00
.03
.08
-.08
.08
5
1.00
.16
.06
.09
.07
.29
6
-
1.00
-.16
-.04
-.02
-.02
-.03
-.19
7
1.00
-.004
-.20
.08
.07
.09
-.02
.05
8
Supervisor, 3 = Work Incentives, 4 = Pay, 5 = Praise &
Factor names: 1 = Evaluation, 2 = Dependable
Job Security, 8 = Administration & Supervision.
Work Location, 6 = Housing & Transportation, 7 =
6
6
In recruiting staff for inservice training,
The name given to each factor by. the re-
well.
agents reported that merit (good work) was often
searcher, assisted by a panel of experts, and the
overlooked or not seriously considered. Although
percent the factor explained were as follows:
other important factors such as one's past academic
evaluation (7.4%), dependable supervisors (5.3%),
record, work performance, years since graduation,
work incentives (2.8%), Pay (2.2%), praise and work
and home district were considered in selecting an
location (1.8%), housing and transportation (1.6%),
agent for further training, agents suggested that for
job security (1.5%) and administration and super-
inservice training to motivate, merit ought to be the
vision (1.3%). Table 3 shows means and standard
most important criterion in the selection process.
deviations for items with factor loadings of .40 or higher.
Other things that agents said were important
Objective Three
for their motivation included adequate and regular
payment of travel, sub: lence and hotel allow-
The objective sought to determine the job
ances; and health insurance. Some agew s felt that
motivational level of the agents. Table 4 indicates
technical staff, irrespective of rank, should receive
the mean and standard deviation of each variable
equal per diem for travel and hotel accomir.1ation.
used to measure motivation. The overall agents'
Per diem for rail travel and hotel expenses in Kenya
scale
mean moti.vational-level score was 3.66 on a
for officers on duty is based on rank. Consequently,
of 1 to 5 where 1 represented the lowest and 5 the
it was difficult for low-ranking agents to stay in the
highest motivation. Within the ranks, the motiva-
same hotels with high-ranking agents even when
tional level was 3.80 for Agricultural Assistants,
working on similar assignments because of reim-
3.43 for Assistant Agricultural Officers, and 3.40 for
bursement problems. However, other agents viewed
An analysis of variance
Agricultural Officers.
per diem based on rank as a good incentive for
indicated that the differences in the motivational
attracting low-ranking agents to move up through
level of these three groups were statistically
This view tended to be held by high-
the ranks
significant. The Thkey's HSD (honestly significant
ranking agents while low-ranking agents tended to
difference) post hoc test was applied to determine
hold the opposing view.
The findings indi-
which groups were different.
cated that Agricultural Assistants were signifi-
Objective Four
cantly different in their motivational level from
Assistant Agricultural Officers and Agricultural
Objective four sought to examine the relation-
Officers but Assistant Agricultural Officers were
ships between independent variables and the de-
Offic-
not significantly different from Agricultural
pendent variable. An examination of intercorrela-
In terms of gender, an F test revealed no
ers.
tions among extraneous independent variables
statistically significant difference between the mo-
revealed very strong association between the agents'
tivational level of males (3.66) and that of females (3.67).
between
age and total years in service (r=.90), and
the agents' rank and formal education (r=.95).
The researcher also conducted group inter-
views. The agents were asked to describe their
The presence of highly correlated independent
highly
* motivation in terms of whether they were
variables, indicated that one variable can be
motivated, motivated, or not motivated. In eleven
explained or predicted by the other.
This
districts, the agents said they were not motivated
multicollinearity limits the size of R2 or incorrectly
while in one district they said they were motivated.
estimates the regression coefficients because of
The district in which the agents described them-
confounding the effects of the independent vari-
selves as motivated had a District Agricultural
ables (Hair, et al. 1992).
Officer who had a reputation of being an effective
and dependable supervisor. The officer had won
To control the undesirable effects of
agents' trust and the praise and admiration of
multicollinearity, the variable less strongly corre-
This particular finding further con-
superiors.
lated with the dependent variable in each pair was
firmed the overall findings of the study which
dropped from further analysis. Thus agents' age
indicated that having dependable supervisors was
while
was dropped in favor of total years in service
he most important factor related to the motivation
formal education was dropped in favor of the
7xtension agents in Kenya's Rift Valley Province.
agents' rank on the job.
The group interviews also revealed that most
Dummy coding of categorical variables such as
related
agents believed their promotions were more
gender, marital status, rank and highest qualifica-
performance.
to years of service than to individual
tion was used before entering these variables into
Therefore, they had little motivation to perform
7
7
Table 3
MEANS AND STANDARD DEVIATIONS FOR VARIABLES
COMPRISING THE EIGHT JOB SATISFACTION FACTORS (N=325)
Mean
SD
Abbreviated Variable Label*
Item
4.18
.82
Being recognized for good work increases my motivation
60
.77
4.14
Evaluation of my work motivates me to work harder
87
.78
4.21
Feedback from my supervisor increases my motivation.
33
.82
4.42
Positive recognition makes me proud to be an agent
45
3.74
1.05
I enjoy meeting my supervisor to discuss my work.
92
3.52
1.14
Effective supervisors praise agents for good performance
61
.77
4.35
My supervisor's feedback gives me confidence in my job
20
.89
4.09
Praise for good performance increases my desire to excel
116
1.30
3.20
My supervisor tends to concentrate more on my mistakes
43
3.63
1.31
I get more negative input than help from my supervisor
28
1.32
2.87
I frequently receive positive recognition for good work
15
1.14
1.81
In Extension most hardworking agents go unrewarded
30
1.14
2.56
I am satisfied with most of the current &tension policies
57
1.10
3.33
My supervisor makes my work more pleasant -
97
1.39
3.28
I have a chance to do things for which I am most qualified
16
1.02
2.10
I work hard mainly to avoid being disciplined
as
3.83
1.01
Evaluating me on work objectives would lower my motivation
37
3.92
1.17
I am more motivated by pay than by the work I do
24
.96
3.94
Higher pay is more important to me than job security
84
2.15
1.02
In extension, pay is the most important thing to me
58
1.10
3.72
Praise has little influence on my work performance
46
1.16
4.06
I deserve little positive recognition for doing my job well
36
1.20
3.79
I prefer working far away from my home area
32
1.11
3.39
I should be praised less frequently for doing my job well
59
1.00
3.69
Being praised:I makes me feel flattered
52
.85
4.11
Good housing increases my motivation to work
80
3.96
.93
Housing has little influence on my job satisfaction
82
.74
4.42
Good housing contributes to favorable work environment
51
.94
4.26
Good housing contributes to my job satisfaction
6
1.05
4.15
Inadequate transport reduces my job effectiveness
76
.79
4.38
Adequate transport gives me job satisfaction
99
1.39
that pays more
3.62
I prefer a secure job that pays less than insecure one
22
.82
4.38
Feeling secure on the job motivates me to work harder.
26
1.00
3.83
performance
Extension administration has little influence on my work
100
1.08
3.55
work
Supervision from my boss has little effect on how I
96
& 100 which were negatively stated were
'Items 24 28, 30, 32, 36, 37, 43, 46, 52, 59, 82, 84, 96,
where 1=Strongly Disagree, 2=DIsagree,
recoded. The items were rated on a scale of 1 to 5
3=Undecided, 4=Agree, 5=Strongly Agree.
TABLE 4
MOTIVATIONAL LEVEL (N=325)
AGENTS' JOB MOTIVATION VARIABLES AND MEAN
Mean
SD
Variable Label'
Item
3.76
1.27
I often think of leaving the Extension service
9
3.79
1.07
Working as an Extension agent is in itself rewarding
38
1.38
3.27
I am highly motivated as an agent
42
.74
4.33
I love my job
64
1.14
3.91
My job is frustrating
102
1.03
4.04
I wish I had chosen a different career
105
1.03
3.67
most
The hours I spend on the job are the ones I enjoy
109
3.80
1.08
be an Extension agent
If I were to choose a career once more, I would rhoose to
113
2.90
1.16
work
While on vacation, I often wish I were back to
114
.27
3.15
for personal growth
In the Extension service, I have many opportunities
115
recoded before calculating the variable means and
' Negative items 9, 102, and 105, were
ote:
and 5 the highest level of motivation).
)
standard deviations (1 represented the lowest
= 5.0
= 1.4
Max
= 0.72
Min
Mean = 3.66
SD
8
8
objectives of the study and are generalizable to
the regression equation. When simultaneous and
stepwise procedures were used to regress motiva-
Extension agents in Kenya's Rift Valley Province.
tion on personal characteristics, only agents' rank
They are as follows:
and years of service were statistically significant (p < .05).
Personal characteristics were not as important
1.
for the agents' motivation as were the job
The two regression procedures were repeated
satisfaction factors.
using the job satisfaction-factor scores plus agents'
rank and years of service. Th e results showed that
Eight job satisfaction factors were important for
2.
only rank and five of the eight job satisfaction
the agents' job satisfaction. In decreasing order
of their importance, these factors were evalua-
factors were statistically significant (p < .05).
tion, dependable supervisors, work incentives,
pay, praise and work location, housing and
Following McCracken's (1991) recommenda-
transportation, job security, and administration
tion for controlling the effects of extraneous inde-
pendent variables, the researcher entered the
and supervision.
Only five of the eight job satisfaction factors
agents' rank on the job into a hierarchical regres-
3.
sion equation first to determine the amount of
were important for the agents' motivation. In
&Creasing order of their importance, these
unique variance it contributed Rank accounted for
factors were dependable supervisors, pay, job
about 7% of the variance in the agents' motivation.
security, evaluation, and administration and
The job satisfaction factors which were statistically
significant in the simultaneous and stepwise re-
supervision.
One cause of agents' frustration and low job
gression equations were then entered into the
4.
motivation, according to the group interviews,
hierarchical regression equation (see Table 5) in
was their perception that merit was being
order of their importance.
ignored, or not seriously considered, in select-
After entering all the factor scores into the
ing candidates for inservice staff training.
On the basis of the questionnaire used to
equation, the final R2 was .55 while adjusted R2 was
5.
.54 indicating that the job satisfaction factors
measure motivation in this study, the agents'
contributed an additional 48% of the variance in
motivational level (3.66) was above the mid-
motivation after accounting for the variance con-
point (2.50) of the motivational scale used in the
tributed by the agents' rank However, the final t
questionnaire
test values indicated that rank, represented by
Over half of the variance in the agents'
6.
Dummy3 and Dummy4, was not statistically sig-
motivation can be explained by five job satisfac-
nificant (p < .05). When motivation was regressed
tion factors.
on the five job satisfaction factors in a stepwise
Job satisfaction and motivation are related but
7.
regression equation omitting rank, an R2 of .54 and
different. This conclusion was illustrated by the
fact that eight latent factors were used to
an adjusted R2 of .53 were obtained.
explain approximately one fourth of the varia-
Conclusions
tion in agents' job satisfaction. However, only
five of these factors were used to explain more
than one halt of the variation j4i their motiva-
On the basis of this study, the researcher made
several conclusions, which are related to the
tional level.
Table 5
REGRESSION OF AGENTS' MOTIVATION ON AGENTS' RANK AND SELECTED JOB
SATISFACTION FACTOR SCORES - HIERARCHICAL ENTRY (N=325)
t
b
Factors
R2Ch
R2
I)
.7025
.38
.48
.068
.068
Dummy3
Rank:
.2876
-1.07
-1.34
.000
Dummy4
.068
.0000
10.96
3.46
.284
Dependable Supervisors
.352
.0000
-7.86
-2.55
.097
.449
Pay
.0000
-5.20
-1.77
Job Security
.061
.510
.0000
4.68
1.51
.029
Staff Evaluation
.539
.0016
-3.19
1.18
.015
.554
Administration and Supervision
36.67
Constant
Adjusted R2 = .544, Model F = 52.96, p<.0001
Note: Standard error = 4.87,
9
job satisfaction. Therefore, raising agents' motiva-
Though generalizable only to agents in Kenya's
8.
tion also increases their job satisfaction.
Rift Valley Province, the findings may be useful
to Extension managers in other parts of the
country, because Kenya's Extension agents
The study supports Herzberg's (1959) fmdings
regarding the importance of good relations with
have similar basic training and terms of service.
one's supervisor, administrative support and good
Implications
supervision, good pay and job security. However,
the study does not support his view that these
In this study
factors cannot satisfy employees.
Since none of the agents' personal character-
eight job satisfaction factors contributed 24% of the
istics were significantly related to their motivation,
variance
job satisfaction.
Extension managers could do a better job of
improving staff motivation by giving less attention
The findings strengthen Vroom's (1964) rec-
to personal characteristics, and more attention to
onunendation that staff performance be assessed
the job satisfaction factors identified in this study.
accurately; based on standards that employees
In 1973, Lawler III reported that in most manufac-
perceive to be fair, achievable, and equal for all.
turing jobs, the best worker produced two to three
The study supports his view that the entire
times as much as the worst worker while in other
incentive system must get administrative backing
jobs, differences were even greater. He indicated
and attention. The findings also agree with the
that in addition to motivation, staff performance
observation made by Adams and Rosenbaum (1962)
was influenced by one's ability and other factors
that treating employees inequitably will lower their
such as mechanical breakdowns and low-quality or
motivation. Finally, the study supports Skinner's
inadequate supply of materials. He pointed out that
(1969) conclusion that identical rewards for all
if little ability was required and people had the
employees are ineffective motivators. To motivate,
same anility, skills, and training, positive staff
rewards must be based on individual performance.
motivation was the most important determinant of
effective job performance.
In the revised conceptual framework of the
study (see Figure 3), the extraneous variables,
This study has shown that for Extension
comprised of personal characteristics, were omitted
agents in Kenya's Rift Valley Province, having
after the study showed that they were not signifi-
dependable supervisors was the most important
cantly related to the agents' motivation. Over half
factor related to motivation. Therefore, investing
of the variance in the motivation of Kenya's Rift
in propor selection and training of Extension
Valley Extension agents can be explained by the
supervisors is the most important step in improving
five job satisfaction factors, which can be manipu-
staff motivation, performance and productivity. In
lated to improve motivation.
addition to having dependable supervisors, staff
motivation can be increased significantly by tying
The difference between dependable supervi-
agents' pay to performance, providing job security,
sors and supervision (see Figure 2 & 3) might not
evaluating agents objectively and having an admin-
be obvious but the two factors are different.
istration and supervision whose primary concerns
Dependable supervisors symbolize a good interper-
are staff productivity and welfare. This study has
sonal relationship between the supervisor and the
also shown that motivation is positively related to
Motivation*
Job Satisfaction
1 Dependable Supervisor
1 Evaluation
2 Pay
2 Dependable Supervisor
3 Job Security
3 Work Incentives
4 Evaluation
4 Pay
5 Administration & Supervision
5 Praise & Work-Location
6 Housing & 1Vansportation
6 Housing & Transportation
7 Praise & Work Location
7 Job Security
8 Work Incentives
8 Administration 4t Supervision
* Only factors 1-5 were significantly related to agents' motivation at .05 alpha level.
Figure 2. The relationship between selected job satisfaction factors and
agents' motivation in Kenya's Rift Valley Province
1 0