this PDF file The Effect of Good Agriculture Practices (GAP) on Soybean Productivity with CobbDouglas Production Function Analysis in Kulon Progo Regency | Purnamasari | Agro Ekonomi 2 PB
220
Agro Ekonomi Vol. 28/No. 2, Desember 2017
THE EFFECT OF GOOD AGRICULTURE PRACTICES (GAP) ON
SOYBEAN PRODUCTIVITY WITH COBB-DOUGLAS PRODUCTION
FUNCTION ANALYSIS IN KULON PROGO REGENCY
Pengaruh ‘Good Agriculture Practices’ (GAP) Terhadap Produktivitas Kedelai
dengan Pendekatan Fungsi Produksi Cobb-Douglas di Kabupaten Kulon Progo
Fitry Purnamasari, Lestari Rahayu Waluyati, Masyhuri
Faculty of Agriculture, Universitas Gadjah Mada
St. Flora-Bulaksumur, Yogyakarta
itrypichenk@gmail.com
Diterima tanggal : 21 Juli 2017 ; Disetujui tanggal : 17 September 2017
ABSTRACT
This study aims to determine the level of adoption of Good Agriculture Practices (GAP)
and the inluence of GAP and other factors of production on soybean productivity. The
number of respondents in this research was 50 farmers taken by random sampling. This
research used proportional parameter test and multiple linear regression analysis with
Ordinary Least Square (OLS) method. This research has been declared valid, reliability,
data have been the normal distribution, free from multicollinearity and heteroscedasticity
problem. The result of the analysis shows that (1) the adoption rate of GAP of soybean
farmers in Kulon Progo Regency is categorized as a high category. Farmers adopted
83,07% of the overall GAP portion of input, land preparation, planting, fertilizing,
crop protection, irrigation, harvesting, and post-harvest. (2) The result of the R2test
shows that 47,8% variation of soybean productivity can be explained by the eight
independent variables and the remaining is explained by variables outside the model.
F test results show that the independent variables together affect the productivity of
soybeans. The result of t test shows that Seed, manure, Gandasil fertilizer, GAP adoption
rate signiicantly positive and NPK fertilizer signiicantly negatively affect soybean
productivity.
Keywords : Cobb-Dauglas, GAP, Productivity, Soybean
INTISARI
Penelitian ini bertujuan untuk mengetahui tingkat adopsi Good Agriculture Practices
(GAP) serta pengaruh GAP dan faktor produksi lainnya terhadap produktivitas kedelai.
Jumlah responden dalam peneltian ini sebanyak 50 orang petani yang diambil secara
acak. Penelitian ini menggunakan pengujian parameter proporsional dan analisis
regresi linear berganda dengan metode Ordinary Least Square (OLS). Penelitian
ini telah dinyatakan valid, reliable, telah berdistribusi normal, bebas dari masalah
multikolinearitas dan heteroskedastisitas, sehingga dapat dilakukan analisis regresi
linear berganda. Hasil analisis menunjukkan bahwa (1) tingkat adopsi GAP petani
kedelai di Kabupaten Kulon Progo tergolong dalam ketegori tinggi. Petani mengadopsi
sebesar 83,07% dari keseluruhan bagian GAP yaitu input, persiapan lahan, penanaman,
pemupukan, perlindungan tanaman, pengairan, panen dan pascapanen. (2) Hasil uji R2
Agro Ekonomi Vol. 28/No. 2, Desember 2017
221
menunjukkan bahwa 47,8% variasi produktivitas kedelai dapat dijelaskan oleh kedelapan
variabel independen dan sisanya dijelaskan oleh variabel diluar model. Hasil Uji F
menunjukkan bahwa variabel independen secara bersama-sama berpengaruh terhadap
produktivitas kedelai. Hasil Uji t menunjukkan bahwa benih, pupuk kandang, pupuk
Gandasil, tingkat adopsi GAP secara signiikan berpengaruh positif dan pupuk NPK
secara signiikan berpengaruh negatif terhadap produktivitas kedelai.
Kata Kunci : Cobb-Dauglas, GAP, Kedelai, Produktivitas
INTRODUCTION
is 32,796 tons), South Sulawesi 6.20%
Soybean which the latin name is
(average production is 38,036 tons), and DI
Glycine max (soybean yellow); Glycinesoja
Yogyakarta 2.26% (average production is
(black soybean) is a versatile plant. One
13,886 tons) (Badan Pusat Statistik, 2016).
of the important food commodities for
Kulon Progo Regency is one of the soy-
Indonesia after rice plant and maize is
producing districts in Yogyakarta besides
soy. The average demand for soybean in
Gunungkidul and Sleman. Soybean crop
Indonesia annually is ± 2.2 million tons
is used as a crop after rice plant. Soybean
of dry seeds. Meanwhile, according to
production is still low.
BPS data in 2015 showed that production
Low soybean production is caused
only reached 963,099 tons, while the rest
by several things, which are: (1) lack of
is imported from other countries.
land allocation deinitively and speciically
Indonesia’s soybean production
intended for soybean production systems;
centers located in seven (7) provinces,
(2) high risk farming of soybean, low
contributing 84% of national soybean
productivity and low income of soybean
production in 2015, and 27 other provinces
farming; (3) the subject of soybean farming
contributing 16%. The largest contribution
is a traditional farmer with a small scale;
was given by Province of East Java at
(4) the adoption of production technology
33.89% (production number is 208,067
is slow; and (5) the data of harvested area
tons), followed by West Nusa Tenggara
is less accurate, often biased and soybean
15.47% (production number is 94,948
production enhancement program is not
tons), and Central Java 11.51% (production
focused on the expansion of new areas
number is 70,629 tons). Four other
(Sumarno and Adie, 2010).
central provinces contributing under
According to Alimmoeso (2008),
10%, that is West Java with the number
quoted by Zakaria (2010) suggested that
is 9.8%(production number is 60,172
the increase in soybean production could
tons), Aceh 5.34% (average production
be done with the expansion of planted area,
222
Agro Ekonomi Vol. 28/No. 2, Desember 2017
the increasing in productivity, production
most important production factors. The
security and institutional strengthening.
relationship between production factors
A good guidelines for the cultivation
(input) and production (output) is usually
of food crops/ Good Agriculture Practices
called as production function or referred
(GAP) was published as a guideline in
to the relationship factor (Isnowati, 2014).
carrying out the cultivation of food crops
Soekartawi (1990), quoted by
correctly and properly, with the expectation
Rahayu (2010) states that the use of
it will be obtain a high productivity, good
production factors such as land, labor,
product quality, maximum profitability,
capital (fertilizers, seeds and medicines)
environmental friendliness and attention
plays an important role in farming activity.
to security aspects, health and welfare
In agricultural development, inaccuracy of
of farmers and sustainable production
number and combination for production
efforts. The legal basis of GAP in Indonesia
factors produce a low production result
for food crops regulated in Ministry of
or high cost of production which makes
Agriculture Regulations Number 48/
farmer income is low.
Permentan/ OT.140/ 10/2006.
The purpose of this study is to
Setiawan et al (2015) in his research
determine the level of GAP adoption
about GAP (Good Agriculture Practice)
on soybean commodity in Kulon Progo
implementation of pepper and the effect
Regency and the effect of GAP and soybean
on pepper productivity has significant
production factors on productivity.
influence accumulatively. Bayramoglu,
et al(2010) EurepGAP certified tomato
METHODS
producers gained higher output and gained
This research used secondary and
2.8 times more net proit than non-certiied
primary data to get the data. Secondary
EurepGAP.
data was a data obtained from literature
Soybean productivity also affected
and a reference document from related
by other factors of production besides GAP.
institutions. Primary data obtained directly
Production factors is known as the term
from farmers used a questionnaire by
inputs, production factors and production
interview. The number of respondents
sacriice. Production factors determinethe
in this research was 50 farmers taken by
size of production, whether it is big or
random sampling. An instrument that was
small. In various experiences indicate
given to the farmer respondents should
that the land as production factors, the
be analyze its validity and reliability in
capital to buy seeds, fertilizer, medicine,
advance used SPSS 16.0, so the data
labor and management aspects are the
will valid and reliable. A measuring
Agro Ekonomi Vol. 28/No. 2, Desember 2017
instrument was said to have good validity
Description:
if it can measure accordingly. Validity
P
(questionnaire) was calculated with the
223
= Percentage of adoption rate of
observations
formula fromproduct moment correlation
P0
= Percentage of adoption rate (at 0.50)
(Pearsons correlation) as follows:
N
= Number of Samples
To determine the factors that affect
Description:
r xy = coeficient of correlation per item
N
= Number of respondents
X
= Score per item / items
Y
= Total Score
the soybean productivity in Kulon
Progo Regency it can use the multiple
regression analysis with Ordinary Least
Square method (OLS). The model use
the equation of Cobb-Douglas function
as follows:
Product Moment Correlation then
Ln Yq = ln α + b1lnx1 + b2lnx2 + b3lnx3 +
compare correlation number with critique
b4lnx4 + b5lnx5 + b6lnx6 + b7lnx7 +
numbers from table correlation r which is
b8lnx8 + μ
its value should be at certain degree (5%).
Description:
If the correlation number is greater than the
YQ = Soybean Productivity (Kg/ Ha)
igures in the table r = 0.632, then the item
a
is declared as valid question.
b1,b2,b3,b4,b5 = regression coeficient
= value of the constant
Reliability test is the consistency
x1
= Land (Ha)
of a measuring instrument that shows the
x2
= Seeds (kg/Ha)
extent to which the measuring instrument
x3
= NPK Fertilizer (kg/Ha)
is trustworthy or reliable when used
x4
= Manure (kg/Ha)
repeatedly. To measure the reliability of
x5
=
Gandasil Fertilizer (gram/ha)
research instrument, it will be carried out
x6
= Pesticides (liters / ha)
x7
= Labor (HOK / ha)
suggested value ofCronbach’s Alphais ≥
x8
= GAP Adoption Rate (score)
0.6.
μ
= mistake factor
statistical tests Cronbach’sAlpha. The
Likert scale is used to measure
the adoption level of GAP, then it use
parameter proportion test as follows.
Classical Assumption Test
To get a good model or qualiiy for
multiple linear regression, model analysis
with classical assumption (normality test,
224
Agro Ekonomi Vol. 28/No. 2, Desember 2017
multicollinearity and heterocedasticity)
t test. The coeficient of determination (R2)
should be done.
describes how much percentage of the total
Normality test analyze use the
variation in the dependent variable could
Jarque-Bera (JB). This test was based on
be explained by the model, the greater the
the coeficient of kurtosis (kurtosis and
R2 the greater inluence in explaining the
coeficient slope or skewness). The formula
dependent variable models. R2 values has
Jarque Bera as follows.
range from 0 to 1, an R2 of 1 means there is a
perfect match, while the value 0 indicates no
relationship between the dependent variable
with the variables explained. Therefore, the
The formula above contain n as sample
size, S describesthe slope and K expresses
the wedge. This test compared the statistics
Jarque-Bera (JB) with a value from Table. If
the value of Jarque-Bera (JB) ≤ Table then
the residual value has normal distribution.
Multicollinearity is a condition
which hasliniear relationship between
independent variables. Multicollinearity
value of R2which is close to 1, means the
closer of its model’s ability to explain the
dependent variable, and otherwise.
F statistic test determine the effect
of simultaneous independent variable on
dependent variable whether it is signiicant
or not. The t-test is used to determine the
effect of independent variable partially
towards dependent variable.
can be seen from the value of tolerance and
the opponent of Variance Inlation Factor
(VIF). If the value of tolerance ≤ 0,10 or
VIF ≥ 10, then there is a multicollinearity
and otherwise (Ghozali, 2012).
Heterocedasticity test conducted
to determine whether there is the same
distribution of the residual variance. If
the probability values in Obs*R-Squared
>critical value (α), then H0is accepted, it
means there is no heterocedasiticity and
otherwise.
RESULTS AND DISCUSSION
Adoption Level of Good Agriculture
Practices (GAP)
In this study, GAP is divided
into inputs, land preparation, planting,
fertilization, plant protection, irrigation,
harvest and post-harvest. Validity test using
product moment correlation coeficient of
Karl Person indicates that the instrument
is valid with r-count value is greater than
the value of r-table Product Value Critical
Moment (0.2306) at the 5% signiicance
Statistic Test
Statistic test includes testing the
coeficient of determination (R2), F test and
level. Reliability testuse Alpha (α) of
Cronbach which show that the value of
Cronbach’s Alpha is greater than 0.6,
Agro Ekonomi Vol. 28/No. 2, Desember 2017
it means the variables are reliable and
consistent (reliable).
225
The Table 2 shows that the average
score for input is 2.65 (88.53%) means that
Respondents were grouped into three
farmers are adopting about 88.53% of the
categories, they are low, moderate and high
overall input. The average score of land
of adoption level based on total score in
preparation is 2.46 (81.52%), planting is
each scope. The number of items which
2.08 (69.33%), fertilizing is 2.43 (80.91%),
were used to measure the level of adoption
plant protection is 2.45 (81.61%), irrigation
of GAP has 57 statement divided into 7
is 2.66 (88.66%), harvest and post harvest
scope. Each statement is adjusted to the
are 2.78 (92.66%). While the total adoption
Regulation of the Minister of Agriculture
of GAP had an average score of 2.49
Number 48/ Permentan/ OT.140/ 10/2006
(83.07%) and classify as high category.
about thegood guidelines for cultivation
Farmers distributiontowards the rate of
of food crops.
GAP adoption can be seen in Table 3 below.
Table 1.The Rate of GAP Adoption for Soybean in Kulon Progo Regency
No.
1.
2.
3.
4.
5.
6.
7.
Total
Scope of Adoption
Input
Land Preparation
Planting
Fertilization
Plant protection
Irrigation
Harvest and Post Harvest
Average Score
2.65
2.46
2.08
2.43
2.45
2.66
2.78
2.49
Percentage (%)
88.53
81.52
69.33
80.91
81.61
88.66
92.66
83.07
Source: Primary Data Analysis, 2017
Table 2. Distribution of Soybean Farmers to Adopt GAP
Low
Scope Adoption
Frequency
Input
Land Preparation
Planting
Fertilization
Plant Protection
Irrigation
Harvest and
Postharvest
Total Level Adoption
of GAP
1
3
12
4
3
3
Category adoption Levels
Moderate
High
Person
Person
Person
Frequency
Frequency
(%)
(%)
(%)
2
4
8
45
90
6
11
22
36
72
24
19
38
19
38
8
8
16
38
76
6
10
20
37
74
6
6
12
41
82
1
2
6
12
43
86
0
0
13
26
37
74
Source: Primary Data Analysis, 2017
226
Agro Ekonomi Vol. 28/No. 2, Desember 2017
Table 3. Analysis Result of GAP Adoption Rate for Soybean
No. ScopeAdoption
1.
Input
2.
Land Preparation
3.
Planting
4.
Fertilization
5.
Plant Protection
6.
Irrigation
7.
Harvest and Postharvest
Total Adoption of GAP
ZCalculate
ZTable
5.657
1.645
3.111
1.645
-1.697
1.645
3.677
1.645
3.394
1.645
4.525
1.645
5.091
1.645
3.394
1.645
Conclusion
Ho rejected
Ho rejected
Ho received
Ho denied
Ho rejected
Ho rejected
Ho rejected
Ho rejected
Criteria
High
High
Average
High
High
High
High
High
Source: Primary Data Analysis, 2017
The adoption rate is measured using
Table 3 show that among 7 scope of
the likert scale of 1 (low), 2 (medium)
GAP, the others 6 which are inputs, land
and 3 (high). Number of statements each
preparation, fertilization, crop protection,
is input 10, land preparation 11, planting
irrigation, harvesting and post harvest have
5, fertilization 11, plant protection 12,
a high adoption rate. While the scope of the
irrigation, harvest and post harvest 5
planting is in moderate category.
statement. Categorization is done by
subtracting the highest score with the
Input
lowest score then divided into three
Supply input was important in
representing each category. Farmers will
soybean cultivation. The inputs in this
be categorized according to the number of
research are seed, labor, and equipment
scores obtained in each scope.
used by the farmers. GAP adoption
Table 2 above show the total value
achievement levels for the input component
of GAP for all its scope, most farmer
are presented in Table 2. In Table 1 shows
respondents are at high category with 37
that the percentage of soybean input usage
people or 74%. While others in moderate
is 88.53% of all assessment aspect. In Table
category have 13 people or 26% and no
2, most farmerswas included in the high
farmer who is in low category of GAP
category (90%), moderate (8%) and low
adoption.
(2%). This is compatible with the statistical
To analyze the hypothesis about GAP
test using parameter proportion analysis
adoption level in Kulon Progo Regency, it
stating that H0 is declined because of Z
should analyze the proportion parameter
score (5.657)> Z table (1.645) which means
Z with a signiicance level of 0.05. The
that the rate of soybean GAP adoptionis
analysis result of GAP adoption rate for
mostly in the high category.
soybean in Kulon Progo Regency can be
seen in Table 3.
The most seeds varieties are used
by farmers is 92% for Grobogan and 8%
Agro Ekonomi Vol. 28/No. 2, Desember 2017
227
for Merapi. Seed pricefor Grobogan is
the irst step in soybean cultivation process.
Rp. 6,000 – 15,000 and for Merapi is Rp.
The average of farmers’ land area is 0.23
14,000. Some farmers use seeds from
hectares. Table 1 shows that the average of
government and input store around their
GAP implementation for land preparation
residence. Farmers respondents use about
section only at 81.52%. The adoption level
75.46 kg/ ha. The suggestion to use quality
of GAP for soybean classiied in the high
seed with the power to grow> 85%, healthy
category (72%), moderate (22%), and low
and clean with the total seed requirement
(6%). Test results hypothesis in Table 3
between 40-60 kg/ ha, depending on seed
states that H0 rejected because of Z score
size, the greater of the seedsize, the more
(3.111)> Ztable (1.645) means that the rate
seeds were used (Direktorat Jenderal
of GAP adoption for soybenmostly is in the
Tanaman Pangan, 2015).Labor value in
high category.
soybean cultivation in Kulon Progo is 93.52
In Kulon Progo, generally farmers
HOK/ Ha. The labor comes from their
plant soybeans in land which is used to
own family and outside of the family. The
plant a rice. According to the rules of the
wages paid for labor is Rp. 30,000 - Rp.
local government, farmers are encouraged
70,000/ person/ day. This labor is used in
to make a crop rotation with the pattern
the process of land preparation, planting,
of rice plant-rice plant-soybeans. Farmers
fertilization, plant protection, irrigation and
usually plant soybean in the ield which
harvest and postharvest.
is used to planted rice plants becauseit
Agricultural tools is the input to
is difficultfor the farmers to cultivate
facilitate the farmers in soybean cultivation.
a new land considering the time and
The tools are hoes, water spray, sickle, and
eficiency. However, the canal for drainage
thresher. Hoe is used in land preparation
is still needed todispose excess water or to
and cultivation process as a tool for
equalize the amount of water. In addition,
weeding, sickle is to crop the stem of
land preparation is also done by using rice
soybean in harvesting and water sprayto
straw as mulch replacement which over
help fertilization and pesticide usage. The
laid evenly on the surface of the land. To
thresher is toseparate betweenshell and
increase the soil fertility, farmers also add
soybean seeds. Some farmers usethresher
compost in rice ields which will be planted
and some of them use manual method.
a soybean.
Land preparation
Planting
Land preparation is the part of
Planting is done after land
soybean cultivation. Land preparation is
preparation. Table 1 shows that farmers
228
Agro Ekonomi Vol. 28/No. 2, Desember 2017
only apply approximately 69.33% of GAP’s
covered with rice straw to keep moisture
component of cultivation. Generally,
and avoid the pests.
farmers are not doing processing. Farmer
planta rice plant directlyin land that has
Fertilization
been used for rice plant. Distribution of
Table 1 shows that the average of
farmers on the adoption rate of GAP in the
farmer only adopt GAP for fertilization
cultivation showed that 38% of farmers
component of 80.91%. Farmers in the
includein higher categories, the moderate
category is included of High (76%),
category also having the same percentage,
Moderate (16%) and Low (8%) category.
while for lower category has 24%. The
The statistical analysisuse proportional
statistical result based on calculation of
parameters test in Table 5.9, states that
parameter proportion in Table 5.9 shows
H0 is declined because Z score (3.677)>
that Ho is accepted because Z score
Z table (1.645) means that the rate of GAP
(-1.697) < Ztable (1.645) means that the
adoption for soybean tegolong mostly in
rate of GAP-SOP adoption for soybeans
the high category.
mostly include in the moderate category.
Farmers are included in the moderate
Farmers included in the low and
category because some farmers do not
moderate category because some farmers
routinely use organic fertilizer and still
do not follow the recommended planting
prioritize the use of inorganic fertilizer.
distance of 40x15 cm and 40x10 cm, they
Fertilization is to increase the fertility of the
only follow the distance of rice planting
soil so that soybean will grow well. Manure
without any land processing. In addition,
is given before planting the crop, at lowering
some farmers use seeds from previous
stage when NPK fertilizer was given, and
products instead of seeds that are certiied
when Gandasil was given. Gandasil is a
or labeled.
foliar fertilizer, which is given to the plant
Soybean planting in Kulon Progo
by dissolved in water and sprayed evenly
2016 was conducted in May-June. The
on the twigs and leaves. There are two kinds
planting schedule is set up by local
of Gandasilwhich is used by farmers, they
government in order not to disrupt the
are Gandasil D for leaves, it is used when
schedule of the next cropping season.
soybeans is in growing and recovery stage.
Generally farmers plant for two days after
The other kind of Gandisil is Gandasil
harvesting. The farmers make a hole in the
B which is used in generative phase, the
ground following the range which is used
meaning of Gandasil B is a fruit. Gandasil
to madefor rice plant. Each hole was given
B is used to support the formation of lower
two soybean seed. After that, the plant is
buds and form carbohydrates in the fruit. The
Agro Ekonomi Vol. 28/No. 2, Desember 2017
229
average use of NPK fertilizer is 107.91 kg/
a season. To prevent the growth of weeds,
ha, compost is 2637.74 kg/ ha and Gandasil
farmers use rice straw to cover the seed in
is 15.19 pack/ ha (100gr/ pack).
land preparation and planting process. Kind
of pesticides which is used include Bassa,
Plant Protection
Most of the farmers have been
Darmabas, Fastac, Extragen, Lannate, and
Marshal.
applying GAP for plants protection
componentwhich is 81.61% (Table 1).
Irrigation
Farmers distributiontowardsadoption rate of
GAP for watering component have
crop protection according to Table 2 is in high
been adopted by 88.66% according to
category 74%, while 20% is for moderate
Table 1. Farmers distribution towards
category and 6% is for low category. This is
GAP adoptionis in high category (82%),
consistent with statistic test results in Table
moderate (12%) and low (6%). While the
4 which indicates that H0 is rejectedbecause
statistical test results in Table 3 concluded
Z score (3.394)> Ztable (1.645), it means
that H0 is rejectedbecause Z score (4.525)>
that the adoption rate of GAP for soybeanis
Ztable (1.645) means that the rate of GAP
mostly in the high category.
adoption of soybean for irrigation part is
Assessment indicators for fertilization
included in high category.
include the use of inorganic fertilizers listed
Soybean is a plant which is tolerant
or recommended by the government, the
of dry conditions. However, soybeans
use of organic fertilizers derived from plants
require water during the growth period.
or animals, the use of appropriate fertilizer
Four stages of phase on soybean plants
quality, time, dose and mode of application,
that require water is in early growth phase,
prioritizing the use of organic fertilizers
during lowering, pod formation and seed
rather than inorganic, spraying fertilizer done
illing (Adisrwanto, 2008).
without leaving Residue and does not cause
pollution and keep the fertilizer in a clean,
Harvest and Post-harvest
safe and closed place. Farmers included in
GAP adoption rate of harvest and
the moderate category means that farmers
post-harvest has been adopted by farmers
do not routinely carry out plant protection in
with the amount of 92.66%. Farmers who
accordance with assessment indicators.
belong to the high category by 86%, while
Crop protection made by farmers is to
moderate category is 12% and low category
control pest and disease. Spraying is done
is 2%. It can be said that most farmers is in
only when there are pests to avoid the risk
high category. This is according to the results
of crop. Spraying usually only done once in
of statistical tests using the proportional
230
Agro Ekonomi Vol. 28/No. 2, Desember 2017
parameter which concluded that H0 is
heteroskedasticity done to get a good
rejected because Z score (5.091)> Z table
model or qualify multiple linear regresi
(1.645). It means that the level of GAP
before analyze with analysis tools Eviews
adoptionfor harvest and post-harvest soybean
9.0. Normality Test Results indicate
crop largely classiied in the high category.
that the Jarque-Bera value is 0.737 with
Soybean plants which are ready to be
probability 0.609> 0.05 (95% conidence
harvested is marked by falling leaves (the
level) which means that the residual
amount of leaves in the crops are about
distribute normally.
5-10%) and 95% yellow pods are yellow.
The analysis of multicolliniearityis
Generally, for Grobogan varieties are
used to determine whether an independent
harvested in less than 80 days. Harvesting
variable in the regression equation is
is done by cutting the stem using a sickle.
not correlated with seeing the value
To threshing, the farmer use the Power
of tolerance and the value of Variance
Treser (threshing machine) and some of
Inlation Factor (VIF).
them use manual method with a wooden
Table 4 shows that the all of the
bat. Afterwards, the beans are cleaned
variables that inluence soybean yield has
and drying for 2-3 days until the moisture
a value of VIF is less than 10, so it can
content of 10-12% and packed in plastic
be concluded that the regression model
bags that can last a long time and are not
is good because there is no correlation
attacked by pests and diseases.
between independent variables problem
or Multicollinearity. Heteroskedasticity
Factors that affect Soybean Productivity
test is to determine whether there is
Classical Assumption Test
the same distribution of the residual
Classical assumption test such
as normality test, multicolinearity and
variance through the analysis of White
Heteroscedastisticity.
Table 4. Test Results of Multicollinearity
No.
1.
2.
3.
4.
5.
6.
7.
8.
Variable
Land (X1)
Seeds (X2)
NPK Fertilizer (X3)
Manure (X4)
Gandasil Fertilizer (X5)
Pesticides (X6)
Labor (X7)
GAP Adoption rate (X8)
Source: Primary Data Analysis, 2017
Value VIF
4.048
1.642
1.843
1.729
1.908
1.710
3.984
1.384
Agro Ekonomi Vol. 28/No. 2, Desember 2017
231
Table 5 show that the value of Prob.
linear regression analysis to determine the
Chi-Squared at Obs*R Squared is 0.074,
factors that affect the soybean productivity
greater than 0.1, it means the regression
in Kulon Progo
model is homokedasticity or in other
words there is no problem assuming non
Statistical Test
heterokedasticity so it can be concluded
Hypothesis testing for factors that
that the data for each independent variables
affect the soybean productivity using
in the model have a homogeneous variance.
multiple linear regression analysis to
The result of classic assumption test
analyze the coeficient of determination
shows that the data was normally distributed,
(R2),T test and test F. Signiicance level
free of problems for multicollinearity and
used in this study is 95% and 90% (α =
heteroscedasticity, so it can do a multiple
5% , α = 10%).
Table 5. Test Results of Heterokedasticity
F-statistic
Obs * R-Squared
2.055
14.310
Prob. F
Prob. Chi-Square
0.063
0.074
Source: Primary Data Analysis, 2017
Table 6. Test Results t Factors Afecting Soybean Productivity
Variable
C
Ln Land (X1)
Ln Seed (X2)
Ln NPK Fertilizer (X3)
Ln Manure (X4)
Ln Gandasil Fertilizer (X5)
Ln Pesticides (X6)
Ln Labor (X7)
Ln GAP Adoption Rate (X8)
R-squared
Adjusted R-squared
F-Statistics
prob (F-statistics)
Coeficient
1.170
0.060ns
0.329***
-0.217***
0.071*
0.142*
0.092ns
0.094ns
0.709*
Prob.
0.536
0.533
0.001
0.003
0.087
0.060
0.160
0.442
0.080
0.598
0.519
7.619
0.000
Source: Primary Data Analysis, 2017
Description:
*** = signiicance level 99%
*
= signiicance level 90%
ns
= non-signiicant
232
Agro Ekonomi Vol. 28/No. 2, Desember 2017
Table 6. shows that the result
area is 0.533 larger than α (0,1) means
ofAdjusted R-Square dis 0.519, which
that any increase in the number of Land
means 51.9% of the variation of soybean
is not signiicantly would reduce soybean
productivity can be explained by the eight
productivity.
independent variables, they are land,
seeds, NPK Fertilizer, manure, Gandasil
Seeds (X2 )
Fertilizer, pesticides, labor and GAP
The result of T-test in Table 6 indicate
adoption rate. The remaining valueis 48.1%
that the seeds is significantly influence
explained by variables outside the model.
towards soybean yield at 99% signiicance
F test results in Table 6 indicate that
level with p-value of 0.001 is smaller than
the independent variable land, seeds, NPK
α (0.01). Regression coeficient value is
Fertilizer, manure, Gandasil Fertilizer,
0.329 has positive effect, it means that
pesticides, labor and GAP adoption rate
the increase of seeds number by 1 percent
influence towards soybean productivity
would increase soybean yield at 0.329
jointlywith the F-statistic equal to 7.619,
percent, assuming other factors remain
with a probability value
Agro Ekonomi Vol. 28/No. 2, Desember 2017
THE EFFECT OF GOOD AGRICULTURE PRACTICES (GAP) ON
SOYBEAN PRODUCTIVITY WITH COBB-DOUGLAS PRODUCTION
FUNCTION ANALYSIS IN KULON PROGO REGENCY
Pengaruh ‘Good Agriculture Practices’ (GAP) Terhadap Produktivitas Kedelai
dengan Pendekatan Fungsi Produksi Cobb-Douglas di Kabupaten Kulon Progo
Fitry Purnamasari, Lestari Rahayu Waluyati, Masyhuri
Faculty of Agriculture, Universitas Gadjah Mada
St. Flora-Bulaksumur, Yogyakarta
itrypichenk@gmail.com
Diterima tanggal : 21 Juli 2017 ; Disetujui tanggal : 17 September 2017
ABSTRACT
This study aims to determine the level of adoption of Good Agriculture Practices (GAP)
and the inluence of GAP and other factors of production on soybean productivity. The
number of respondents in this research was 50 farmers taken by random sampling. This
research used proportional parameter test and multiple linear regression analysis with
Ordinary Least Square (OLS) method. This research has been declared valid, reliability,
data have been the normal distribution, free from multicollinearity and heteroscedasticity
problem. The result of the analysis shows that (1) the adoption rate of GAP of soybean
farmers in Kulon Progo Regency is categorized as a high category. Farmers adopted
83,07% of the overall GAP portion of input, land preparation, planting, fertilizing,
crop protection, irrigation, harvesting, and post-harvest. (2) The result of the R2test
shows that 47,8% variation of soybean productivity can be explained by the eight
independent variables and the remaining is explained by variables outside the model.
F test results show that the independent variables together affect the productivity of
soybeans. The result of t test shows that Seed, manure, Gandasil fertilizer, GAP adoption
rate signiicantly positive and NPK fertilizer signiicantly negatively affect soybean
productivity.
Keywords : Cobb-Dauglas, GAP, Productivity, Soybean
INTISARI
Penelitian ini bertujuan untuk mengetahui tingkat adopsi Good Agriculture Practices
(GAP) serta pengaruh GAP dan faktor produksi lainnya terhadap produktivitas kedelai.
Jumlah responden dalam peneltian ini sebanyak 50 orang petani yang diambil secara
acak. Penelitian ini menggunakan pengujian parameter proporsional dan analisis
regresi linear berganda dengan metode Ordinary Least Square (OLS). Penelitian
ini telah dinyatakan valid, reliable, telah berdistribusi normal, bebas dari masalah
multikolinearitas dan heteroskedastisitas, sehingga dapat dilakukan analisis regresi
linear berganda. Hasil analisis menunjukkan bahwa (1) tingkat adopsi GAP petani
kedelai di Kabupaten Kulon Progo tergolong dalam ketegori tinggi. Petani mengadopsi
sebesar 83,07% dari keseluruhan bagian GAP yaitu input, persiapan lahan, penanaman,
pemupukan, perlindungan tanaman, pengairan, panen dan pascapanen. (2) Hasil uji R2
Agro Ekonomi Vol. 28/No. 2, Desember 2017
221
menunjukkan bahwa 47,8% variasi produktivitas kedelai dapat dijelaskan oleh kedelapan
variabel independen dan sisanya dijelaskan oleh variabel diluar model. Hasil Uji F
menunjukkan bahwa variabel independen secara bersama-sama berpengaruh terhadap
produktivitas kedelai. Hasil Uji t menunjukkan bahwa benih, pupuk kandang, pupuk
Gandasil, tingkat adopsi GAP secara signiikan berpengaruh positif dan pupuk NPK
secara signiikan berpengaruh negatif terhadap produktivitas kedelai.
Kata Kunci : Cobb-Dauglas, GAP, Kedelai, Produktivitas
INTRODUCTION
is 32,796 tons), South Sulawesi 6.20%
Soybean which the latin name is
(average production is 38,036 tons), and DI
Glycine max (soybean yellow); Glycinesoja
Yogyakarta 2.26% (average production is
(black soybean) is a versatile plant. One
13,886 tons) (Badan Pusat Statistik, 2016).
of the important food commodities for
Kulon Progo Regency is one of the soy-
Indonesia after rice plant and maize is
producing districts in Yogyakarta besides
soy. The average demand for soybean in
Gunungkidul and Sleman. Soybean crop
Indonesia annually is ± 2.2 million tons
is used as a crop after rice plant. Soybean
of dry seeds. Meanwhile, according to
production is still low.
BPS data in 2015 showed that production
Low soybean production is caused
only reached 963,099 tons, while the rest
by several things, which are: (1) lack of
is imported from other countries.
land allocation deinitively and speciically
Indonesia’s soybean production
intended for soybean production systems;
centers located in seven (7) provinces,
(2) high risk farming of soybean, low
contributing 84% of national soybean
productivity and low income of soybean
production in 2015, and 27 other provinces
farming; (3) the subject of soybean farming
contributing 16%. The largest contribution
is a traditional farmer with a small scale;
was given by Province of East Java at
(4) the adoption of production technology
33.89% (production number is 208,067
is slow; and (5) the data of harvested area
tons), followed by West Nusa Tenggara
is less accurate, often biased and soybean
15.47% (production number is 94,948
production enhancement program is not
tons), and Central Java 11.51% (production
focused on the expansion of new areas
number is 70,629 tons). Four other
(Sumarno and Adie, 2010).
central provinces contributing under
According to Alimmoeso (2008),
10%, that is West Java with the number
quoted by Zakaria (2010) suggested that
is 9.8%(production number is 60,172
the increase in soybean production could
tons), Aceh 5.34% (average production
be done with the expansion of planted area,
222
Agro Ekonomi Vol. 28/No. 2, Desember 2017
the increasing in productivity, production
most important production factors. The
security and institutional strengthening.
relationship between production factors
A good guidelines for the cultivation
(input) and production (output) is usually
of food crops/ Good Agriculture Practices
called as production function or referred
(GAP) was published as a guideline in
to the relationship factor (Isnowati, 2014).
carrying out the cultivation of food crops
Soekartawi (1990), quoted by
correctly and properly, with the expectation
Rahayu (2010) states that the use of
it will be obtain a high productivity, good
production factors such as land, labor,
product quality, maximum profitability,
capital (fertilizers, seeds and medicines)
environmental friendliness and attention
plays an important role in farming activity.
to security aspects, health and welfare
In agricultural development, inaccuracy of
of farmers and sustainable production
number and combination for production
efforts. The legal basis of GAP in Indonesia
factors produce a low production result
for food crops regulated in Ministry of
or high cost of production which makes
Agriculture Regulations Number 48/
farmer income is low.
Permentan/ OT.140/ 10/2006.
The purpose of this study is to
Setiawan et al (2015) in his research
determine the level of GAP adoption
about GAP (Good Agriculture Practice)
on soybean commodity in Kulon Progo
implementation of pepper and the effect
Regency and the effect of GAP and soybean
on pepper productivity has significant
production factors on productivity.
influence accumulatively. Bayramoglu,
et al(2010) EurepGAP certified tomato
METHODS
producers gained higher output and gained
This research used secondary and
2.8 times more net proit than non-certiied
primary data to get the data. Secondary
EurepGAP.
data was a data obtained from literature
Soybean productivity also affected
and a reference document from related
by other factors of production besides GAP.
institutions. Primary data obtained directly
Production factors is known as the term
from farmers used a questionnaire by
inputs, production factors and production
interview. The number of respondents
sacriice. Production factors determinethe
in this research was 50 farmers taken by
size of production, whether it is big or
random sampling. An instrument that was
small. In various experiences indicate
given to the farmer respondents should
that the land as production factors, the
be analyze its validity and reliability in
capital to buy seeds, fertilizer, medicine,
advance used SPSS 16.0, so the data
labor and management aspects are the
will valid and reliable. A measuring
Agro Ekonomi Vol. 28/No. 2, Desember 2017
instrument was said to have good validity
Description:
if it can measure accordingly. Validity
P
(questionnaire) was calculated with the
223
= Percentage of adoption rate of
observations
formula fromproduct moment correlation
P0
= Percentage of adoption rate (at 0.50)
(Pearsons correlation) as follows:
N
= Number of Samples
To determine the factors that affect
Description:
r xy = coeficient of correlation per item
N
= Number of respondents
X
= Score per item / items
Y
= Total Score
the soybean productivity in Kulon
Progo Regency it can use the multiple
regression analysis with Ordinary Least
Square method (OLS). The model use
the equation of Cobb-Douglas function
as follows:
Product Moment Correlation then
Ln Yq = ln α + b1lnx1 + b2lnx2 + b3lnx3 +
compare correlation number with critique
b4lnx4 + b5lnx5 + b6lnx6 + b7lnx7 +
numbers from table correlation r which is
b8lnx8 + μ
its value should be at certain degree (5%).
Description:
If the correlation number is greater than the
YQ = Soybean Productivity (Kg/ Ha)
igures in the table r = 0.632, then the item
a
is declared as valid question.
b1,b2,b3,b4,b5 = regression coeficient
= value of the constant
Reliability test is the consistency
x1
= Land (Ha)
of a measuring instrument that shows the
x2
= Seeds (kg/Ha)
extent to which the measuring instrument
x3
= NPK Fertilizer (kg/Ha)
is trustworthy or reliable when used
x4
= Manure (kg/Ha)
repeatedly. To measure the reliability of
x5
=
Gandasil Fertilizer (gram/ha)
research instrument, it will be carried out
x6
= Pesticides (liters / ha)
x7
= Labor (HOK / ha)
suggested value ofCronbach’s Alphais ≥
x8
= GAP Adoption Rate (score)
0.6.
μ
= mistake factor
statistical tests Cronbach’sAlpha. The
Likert scale is used to measure
the adoption level of GAP, then it use
parameter proportion test as follows.
Classical Assumption Test
To get a good model or qualiiy for
multiple linear regression, model analysis
with classical assumption (normality test,
224
Agro Ekonomi Vol. 28/No. 2, Desember 2017
multicollinearity and heterocedasticity)
t test. The coeficient of determination (R2)
should be done.
describes how much percentage of the total
Normality test analyze use the
variation in the dependent variable could
Jarque-Bera (JB). This test was based on
be explained by the model, the greater the
the coeficient of kurtosis (kurtosis and
R2 the greater inluence in explaining the
coeficient slope or skewness). The formula
dependent variable models. R2 values has
Jarque Bera as follows.
range from 0 to 1, an R2 of 1 means there is a
perfect match, while the value 0 indicates no
relationship between the dependent variable
with the variables explained. Therefore, the
The formula above contain n as sample
size, S describesthe slope and K expresses
the wedge. This test compared the statistics
Jarque-Bera (JB) with a value from Table. If
the value of Jarque-Bera (JB) ≤ Table then
the residual value has normal distribution.
Multicollinearity is a condition
which hasliniear relationship between
independent variables. Multicollinearity
value of R2which is close to 1, means the
closer of its model’s ability to explain the
dependent variable, and otherwise.
F statistic test determine the effect
of simultaneous independent variable on
dependent variable whether it is signiicant
or not. The t-test is used to determine the
effect of independent variable partially
towards dependent variable.
can be seen from the value of tolerance and
the opponent of Variance Inlation Factor
(VIF). If the value of tolerance ≤ 0,10 or
VIF ≥ 10, then there is a multicollinearity
and otherwise (Ghozali, 2012).
Heterocedasticity test conducted
to determine whether there is the same
distribution of the residual variance. If
the probability values in Obs*R-Squared
>critical value (α), then H0is accepted, it
means there is no heterocedasiticity and
otherwise.
RESULTS AND DISCUSSION
Adoption Level of Good Agriculture
Practices (GAP)
In this study, GAP is divided
into inputs, land preparation, planting,
fertilization, plant protection, irrigation,
harvest and post-harvest. Validity test using
product moment correlation coeficient of
Karl Person indicates that the instrument
is valid with r-count value is greater than
the value of r-table Product Value Critical
Moment (0.2306) at the 5% signiicance
Statistic Test
Statistic test includes testing the
coeficient of determination (R2), F test and
level. Reliability testuse Alpha (α) of
Cronbach which show that the value of
Cronbach’s Alpha is greater than 0.6,
Agro Ekonomi Vol. 28/No. 2, Desember 2017
it means the variables are reliable and
consistent (reliable).
225
The Table 2 shows that the average
score for input is 2.65 (88.53%) means that
Respondents were grouped into three
farmers are adopting about 88.53% of the
categories, they are low, moderate and high
overall input. The average score of land
of adoption level based on total score in
preparation is 2.46 (81.52%), planting is
each scope. The number of items which
2.08 (69.33%), fertilizing is 2.43 (80.91%),
were used to measure the level of adoption
plant protection is 2.45 (81.61%), irrigation
of GAP has 57 statement divided into 7
is 2.66 (88.66%), harvest and post harvest
scope. Each statement is adjusted to the
are 2.78 (92.66%). While the total adoption
Regulation of the Minister of Agriculture
of GAP had an average score of 2.49
Number 48/ Permentan/ OT.140/ 10/2006
(83.07%) and classify as high category.
about thegood guidelines for cultivation
Farmers distributiontowards the rate of
of food crops.
GAP adoption can be seen in Table 3 below.
Table 1.The Rate of GAP Adoption for Soybean in Kulon Progo Regency
No.
1.
2.
3.
4.
5.
6.
7.
Total
Scope of Adoption
Input
Land Preparation
Planting
Fertilization
Plant protection
Irrigation
Harvest and Post Harvest
Average Score
2.65
2.46
2.08
2.43
2.45
2.66
2.78
2.49
Percentage (%)
88.53
81.52
69.33
80.91
81.61
88.66
92.66
83.07
Source: Primary Data Analysis, 2017
Table 2. Distribution of Soybean Farmers to Adopt GAP
Low
Scope Adoption
Frequency
Input
Land Preparation
Planting
Fertilization
Plant Protection
Irrigation
Harvest and
Postharvest
Total Level Adoption
of GAP
1
3
12
4
3
3
Category adoption Levels
Moderate
High
Person
Person
Person
Frequency
Frequency
(%)
(%)
(%)
2
4
8
45
90
6
11
22
36
72
24
19
38
19
38
8
8
16
38
76
6
10
20
37
74
6
6
12
41
82
1
2
6
12
43
86
0
0
13
26
37
74
Source: Primary Data Analysis, 2017
226
Agro Ekonomi Vol. 28/No. 2, Desember 2017
Table 3. Analysis Result of GAP Adoption Rate for Soybean
No. ScopeAdoption
1.
Input
2.
Land Preparation
3.
Planting
4.
Fertilization
5.
Plant Protection
6.
Irrigation
7.
Harvest and Postharvest
Total Adoption of GAP
ZCalculate
ZTable
5.657
1.645
3.111
1.645
-1.697
1.645
3.677
1.645
3.394
1.645
4.525
1.645
5.091
1.645
3.394
1.645
Conclusion
Ho rejected
Ho rejected
Ho received
Ho denied
Ho rejected
Ho rejected
Ho rejected
Ho rejected
Criteria
High
High
Average
High
High
High
High
High
Source: Primary Data Analysis, 2017
The adoption rate is measured using
Table 3 show that among 7 scope of
the likert scale of 1 (low), 2 (medium)
GAP, the others 6 which are inputs, land
and 3 (high). Number of statements each
preparation, fertilization, crop protection,
is input 10, land preparation 11, planting
irrigation, harvesting and post harvest have
5, fertilization 11, plant protection 12,
a high adoption rate. While the scope of the
irrigation, harvest and post harvest 5
planting is in moderate category.
statement. Categorization is done by
subtracting the highest score with the
Input
lowest score then divided into three
Supply input was important in
representing each category. Farmers will
soybean cultivation. The inputs in this
be categorized according to the number of
research are seed, labor, and equipment
scores obtained in each scope.
used by the farmers. GAP adoption
Table 2 above show the total value
achievement levels for the input component
of GAP for all its scope, most farmer
are presented in Table 2. In Table 1 shows
respondents are at high category with 37
that the percentage of soybean input usage
people or 74%. While others in moderate
is 88.53% of all assessment aspect. In Table
category have 13 people or 26% and no
2, most farmerswas included in the high
farmer who is in low category of GAP
category (90%), moderate (8%) and low
adoption.
(2%). This is compatible with the statistical
To analyze the hypothesis about GAP
test using parameter proportion analysis
adoption level in Kulon Progo Regency, it
stating that H0 is declined because of Z
should analyze the proportion parameter
score (5.657)> Z table (1.645) which means
Z with a signiicance level of 0.05. The
that the rate of soybean GAP adoptionis
analysis result of GAP adoption rate for
mostly in the high category.
soybean in Kulon Progo Regency can be
seen in Table 3.
The most seeds varieties are used
by farmers is 92% for Grobogan and 8%
Agro Ekonomi Vol. 28/No. 2, Desember 2017
227
for Merapi. Seed pricefor Grobogan is
the irst step in soybean cultivation process.
Rp. 6,000 – 15,000 and for Merapi is Rp.
The average of farmers’ land area is 0.23
14,000. Some farmers use seeds from
hectares. Table 1 shows that the average of
government and input store around their
GAP implementation for land preparation
residence. Farmers respondents use about
section only at 81.52%. The adoption level
75.46 kg/ ha. The suggestion to use quality
of GAP for soybean classiied in the high
seed with the power to grow> 85%, healthy
category (72%), moderate (22%), and low
and clean with the total seed requirement
(6%). Test results hypothesis in Table 3
between 40-60 kg/ ha, depending on seed
states that H0 rejected because of Z score
size, the greater of the seedsize, the more
(3.111)> Ztable (1.645) means that the rate
seeds were used (Direktorat Jenderal
of GAP adoption for soybenmostly is in the
Tanaman Pangan, 2015).Labor value in
high category.
soybean cultivation in Kulon Progo is 93.52
In Kulon Progo, generally farmers
HOK/ Ha. The labor comes from their
plant soybeans in land which is used to
own family and outside of the family. The
plant a rice. According to the rules of the
wages paid for labor is Rp. 30,000 - Rp.
local government, farmers are encouraged
70,000/ person/ day. This labor is used in
to make a crop rotation with the pattern
the process of land preparation, planting,
of rice plant-rice plant-soybeans. Farmers
fertilization, plant protection, irrigation and
usually plant soybean in the ield which
harvest and postharvest.
is used to planted rice plants becauseit
Agricultural tools is the input to
is difficultfor the farmers to cultivate
facilitate the farmers in soybean cultivation.
a new land considering the time and
The tools are hoes, water spray, sickle, and
eficiency. However, the canal for drainage
thresher. Hoe is used in land preparation
is still needed todispose excess water or to
and cultivation process as a tool for
equalize the amount of water. In addition,
weeding, sickle is to crop the stem of
land preparation is also done by using rice
soybean in harvesting and water sprayto
straw as mulch replacement which over
help fertilization and pesticide usage. The
laid evenly on the surface of the land. To
thresher is toseparate betweenshell and
increase the soil fertility, farmers also add
soybean seeds. Some farmers usethresher
compost in rice ields which will be planted
and some of them use manual method.
a soybean.
Land preparation
Planting
Land preparation is the part of
Planting is done after land
soybean cultivation. Land preparation is
preparation. Table 1 shows that farmers
228
Agro Ekonomi Vol. 28/No. 2, Desember 2017
only apply approximately 69.33% of GAP’s
covered with rice straw to keep moisture
component of cultivation. Generally,
and avoid the pests.
farmers are not doing processing. Farmer
planta rice plant directlyin land that has
Fertilization
been used for rice plant. Distribution of
Table 1 shows that the average of
farmers on the adoption rate of GAP in the
farmer only adopt GAP for fertilization
cultivation showed that 38% of farmers
component of 80.91%. Farmers in the
includein higher categories, the moderate
category is included of High (76%),
category also having the same percentage,
Moderate (16%) and Low (8%) category.
while for lower category has 24%. The
The statistical analysisuse proportional
statistical result based on calculation of
parameters test in Table 5.9, states that
parameter proportion in Table 5.9 shows
H0 is declined because Z score (3.677)>
that Ho is accepted because Z score
Z table (1.645) means that the rate of GAP
(-1.697) < Ztable (1.645) means that the
adoption for soybean tegolong mostly in
rate of GAP-SOP adoption for soybeans
the high category.
mostly include in the moderate category.
Farmers are included in the moderate
Farmers included in the low and
category because some farmers do not
moderate category because some farmers
routinely use organic fertilizer and still
do not follow the recommended planting
prioritize the use of inorganic fertilizer.
distance of 40x15 cm and 40x10 cm, they
Fertilization is to increase the fertility of the
only follow the distance of rice planting
soil so that soybean will grow well. Manure
without any land processing. In addition,
is given before planting the crop, at lowering
some farmers use seeds from previous
stage when NPK fertilizer was given, and
products instead of seeds that are certiied
when Gandasil was given. Gandasil is a
or labeled.
foliar fertilizer, which is given to the plant
Soybean planting in Kulon Progo
by dissolved in water and sprayed evenly
2016 was conducted in May-June. The
on the twigs and leaves. There are two kinds
planting schedule is set up by local
of Gandasilwhich is used by farmers, they
government in order not to disrupt the
are Gandasil D for leaves, it is used when
schedule of the next cropping season.
soybeans is in growing and recovery stage.
Generally farmers plant for two days after
The other kind of Gandisil is Gandasil
harvesting. The farmers make a hole in the
B which is used in generative phase, the
ground following the range which is used
meaning of Gandasil B is a fruit. Gandasil
to madefor rice plant. Each hole was given
B is used to support the formation of lower
two soybean seed. After that, the plant is
buds and form carbohydrates in the fruit. The
Agro Ekonomi Vol. 28/No. 2, Desember 2017
229
average use of NPK fertilizer is 107.91 kg/
a season. To prevent the growth of weeds,
ha, compost is 2637.74 kg/ ha and Gandasil
farmers use rice straw to cover the seed in
is 15.19 pack/ ha (100gr/ pack).
land preparation and planting process. Kind
of pesticides which is used include Bassa,
Plant Protection
Most of the farmers have been
Darmabas, Fastac, Extragen, Lannate, and
Marshal.
applying GAP for plants protection
componentwhich is 81.61% (Table 1).
Irrigation
Farmers distributiontowardsadoption rate of
GAP for watering component have
crop protection according to Table 2 is in high
been adopted by 88.66% according to
category 74%, while 20% is for moderate
Table 1. Farmers distribution towards
category and 6% is for low category. This is
GAP adoptionis in high category (82%),
consistent with statistic test results in Table
moderate (12%) and low (6%). While the
4 which indicates that H0 is rejectedbecause
statistical test results in Table 3 concluded
Z score (3.394)> Ztable (1.645), it means
that H0 is rejectedbecause Z score (4.525)>
that the adoption rate of GAP for soybeanis
Ztable (1.645) means that the rate of GAP
mostly in the high category.
adoption of soybean for irrigation part is
Assessment indicators for fertilization
included in high category.
include the use of inorganic fertilizers listed
Soybean is a plant which is tolerant
or recommended by the government, the
of dry conditions. However, soybeans
use of organic fertilizers derived from plants
require water during the growth period.
or animals, the use of appropriate fertilizer
Four stages of phase on soybean plants
quality, time, dose and mode of application,
that require water is in early growth phase,
prioritizing the use of organic fertilizers
during lowering, pod formation and seed
rather than inorganic, spraying fertilizer done
illing (Adisrwanto, 2008).
without leaving Residue and does not cause
pollution and keep the fertilizer in a clean,
Harvest and Post-harvest
safe and closed place. Farmers included in
GAP adoption rate of harvest and
the moderate category means that farmers
post-harvest has been adopted by farmers
do not routinely carry out plant protection in
with the amount of 92.66%. Farmers who
accordance with assessment indicators.
belong to the high category by 86%, while
Crop protection made by farmers is to
moderate category is 12% and low category
control pest and disease. Spraying is done
is 2%. It can be said that most farmers is in
only when there are pests to avoid the risk
high category. This is according to the results
of crop. Spraying usually only done once in
of statistical tests using the proportional
230
Agro Ekonomi Vol. 28/No. 2, Desember 2017
parameter which concluded that H0 is
heteroskedasticity done to get a good
rejected because Z score (5.091)> Z table
model or qualify multiple linear regresi
(1.645). It means that the level of GAP
before analyze with analysis tools Eviews
adoptionfor harvest and post-harvest soybean
9.0. Normality Test Results indicate
crop largely classiied in the high category.
that the Jarque-Bera value is 0.737 with
Soybean plants which are ready to be
probability 0.609> 0.05 (95% conidence
harvested is marked by falling leaves (the
level) which means that the residual
amount of leaves in the crops are about
distribute normally.
5-10%) and 95% yellow pods are yellow.
The analysis of multicolliniearityis
Generally, for Grobogan varieties are
used to determine whether an independent
harvested in less than 80 days. Harvesting
variable in the regression equation is
is done by cutting the stem using a sickle.
not correlated with seeing the value
To threshing, the farmer use the Power
of tolerance and the value of Variance
Treser (threshing machine) and some of
Inlation Factor (VIF).
them use manual method with a wooden
Table 4 shows that the all of the
bat. Afterwards, the beans are cleaned
variables that inluence soybean yield has
and drying for 2-3 days until the moisture
a value of VIF is less than 10, so it can
content of 10-12% and packed in plastic
be concluded that the regression model
bags that can last a long time and are not
is good because there is no correlation
attacked by pests and diseases.
between independent variables problem
or Multicollinearity. Heteroskedasticity
Factors that affect Soybean Productivity
test is to determine whether there is
Classical Assumption Test
the same distribution of the residual
Classical assumption test such
as normality test, multicolinearity and
variance through the analysis of White
Heteroscedastisticity.
Table 4. Test Results of Multicollinearity
No.
1.
2.
3.
4.
5.
6.
7.
8.
Variable
Land (X1)
Seeds (X2)
NPK Fertilizer (X3)
Manure (X4)
Gandasil Fertilizer (X5)
Pesticides (X6)
Labor (X7)
GAP Adoption rate (X8)
Source: Primary Data Analysis, 2017
Value VIF
4.048
1.642
1.843
1.729
1.908
1.710
3.984
1.384
Agro Ekonomi Vol. 28/No. 2, Desember 2017
231
Table 5 show that the value of Prob.
linear regression analysis to determine the
Chi-Squared at Obs*R Squared is 0.074,
factors that affect the soybean productivity
greater than 0.1, it means the regression
in Kulon Progo
model is homokedasticity or in other
words there is no problem assuming non
Statistical Test
heterokedasticity so it can be concluded
Hypothesis testing for factors that
that the data for each independent variables
affect the soybean productivity using
in the model have a homogeneous variance.
multiple linear regression analysis to
The result of classic assumption test
analyze the coeficient of determination
shows that the data was normally distributed,
(R2),T test and test F. Signiicance level
free of problems for multicollinearity and
used in this study is 95% and 90% (α =
heteroscedasticity, so it can do a multiple
5% , α = 10%).
Table 5. Test Results of Heterokedasticity
F-statistic
Obs * R-Squared
2.055
14.310
Prob. F
Prob. Chi-Square
0.063
0.074
Source: Primary Data Analysis, 2017
Table 6. Test Results t Factors Afecting Soybean Productivity
Variable
C
Ln Land (X1)
Ln Seed (X2)
Ln NPK Fertilizer (X3)
Ln Manure (X4)
Ln Gandasil Fertilizer (X5)
Ln Pesticides (X6)
Ln Labor (X7)
Ln GAP Adoption Rate (X8)
R-squared
Adjusted R-squared
F-Statistics
prob (F-statistics)
Coeficient
1.170
0.060ns
0.329***
-0.217***
0.071*
0.142*
0.092ns
0.094ns
0.709*
Prob.
0.536
0.533
0.001
0.003
0.087
0.060
0.160
0.442
0.080
0.598
0.519
7.619
0.000
Source: Primary Data Analysis, 2017
Description:
*** = signiicance level 99%
*
= signiicance level 90%
ns
= non-signiicant
232
Agro Ekonomi Vol. 28/No. 2, Desember 2017
Table 6. shows that the result
area is 0.533 larger than α (0,1) means
ofAdjusted R-Square dis 0.519, which
that any increase in the number of Land
means 51.9% of the variation of soybean
is not signiicantly would reduce soybean
productivity can be explained by the eight
productivity.
independent variables, they are land,
seeds, NPK Fertilizer, manure, Gandasil
Seeds (X2 )
Fertilizer, pesticides, labor and GAP
The result of T-test in Table 6 indicate
adoption rate. The remaining valueis 48.1%
that the seeds is significantly influence
explained by variables outside the model.
towards soybean yield at 99% signiicance
F test results in Table 6 indicate that
level with p-value of 0.001 is smaller than
the independent variable land, seeds, NPK
α (0.01). Regression coeficient value is
Fertilizer, manure, Gandasil Fertilizer,
0.329 has positive effect, it means that
pesticides, labor and GAP adoption rate
the increase of seeds number by 1 percent
influence towards soybean productivity
would increase soybean yield at 0.329
jointlywith the F-statistic equal to 7.619,
percent, assuming other factors remain
with a probability value