Farmer’s considerations of grain quality based on government policy: Hutabayu Raja District, Simalungun Regency, North Sumatera Province

Based on rice policy, this study shows that grain quality is the most important economically for farmers when selling rice. The study examines factors that influence farmers’ consideration of grain quality when selling GKP. Using a questionnaire, data were collected from 150 randomly selected farmers. Binary logistic regression model and descriptive statistics were used to analyze the data. Binary logit results show that variables such as age, area, and GKP price increase the likelihood that farmers consider grain quality when selling GK. On the other hand, increasing farmers’ education and experience levels has a negative impact on their ability to consider grain quality before selling GKP.


Introduction
One of the objectives of Minister of Trade Regulation No.24 of 2020 is to ensure food availability and achieve food security.To accomplish this, the government has implemented the government purchasing price policy (HPP) for grain or rice.According to Minister of Trade Regulation No.24 of 2020, the government has set specific quality standards for purchasing dried harvest paddy (GKP) and milled dry grain (GKG).These standards include maximum water content for GKP is 25%, and maximum vacuum/impurity content is 10%; maximum water content for GKG is 14%, and maximum empty/impurity content is 3%.Farmers who wish to sell their grain, whether in GKP or GKG form, must meet these quality requirements.Failure to meet these standards will result in a lower price for the grain compared to the HPP set by the government [1].
Previous research has shown that a farmer's decision to sell their harvest is greatly influenced by the quality of the grain.Grain quality is influenced by various factors, including physiological and physicochemical properties, as well as environmental conditions [2].It has been observed that the value of rice increases when there is minimal grain damage during both harvest and post-harvest stages [3].Additionally, factors such as planting methods, harvesting techniques, post-harvest handling, storage management, and transportation methods also impact the quality of grain [4].Moreover, the quality of grain sold by farmers is determined by factors like agroecological zones, production systems, and preharvest activities [5].Furthermore, hot and dry conditions during grain formation, along with the availability of nutrients like nitrogen and sulfur, as well as effective disease control, also contribute to the overall quality of grain [6].
The following study does not consider factors that influence grain quality when selling products such [7] stated that factors that influence farmers in selling grain include agricultural training, contribution to the organization, number of livestock, of the head of the household, farmed land, availability to irrigation, and cost of seeds and [8] found factors that influence agricultural sales performance include geographic location, land area, grain business expertise, farming background, grain storage practices, and application of crop insurance, and [9] demonstrates that grain quality in Ethiopia is influenced by grain moisture content, temperature, initial conditions, insects, pests and fungi [10].Studies on factors that influence the quality of grain/rice from a cultivation perspective include [11] and in terms of the application of shade, the application of N to panicles on the application of rice quality and [12] stated that the chlorophyll index at heading (CIG0) is an acceptable indicator of grain quality in rice.The indicator of rice grain quality uses the percentage of several kinds of imperfect grains and grain protein content (GPC).However, only a small number of studies have analysed the factors that influence farmers' decisions to sell grain based on its quality.
To address the identified gap, the objective of this study was to examine the factors influencing farmers' consideration of grain quality when selling grain in Hutabayu Raja district.Specifically, we hypothesized that age, education, farmer experience, land area, and GKP price would impact the extent to which farmers considering grain quality.We predicted that higher levels of age, education, farmer experience, land area, and GKP price would increase the likelihood of farmers considering grain quality when selling GKP.Therefore, this study seeks to answer the following key research question: What factors influence farmers' consideration of grain quality when selling GKP?

Studi area
The study was conducted at Hutabayu Raja District, Simalungun Regency, North Sumatera Province, Indonesia.Geographically, the district is located between coordinates 2⁰53'-3⁰06'N and 99⁰14' -99⁰21' E, (Figure 1).It has an area of 191.43 square kilometers.According to the Hutabayu Central Statistics Office [13], the district has a total population of 35,426, including 17,354 males and 17,892 females.The wet rice harvest area in 2019 is 7,013.5 hectares with a yield of 39,584 tons and the corn harvest area is 2,165 tons.0.5 hectares with an output of 12,212 tons.

Data collection and sampling
Prior to data collection, a pre-survey was conducted to Agriculture office of Simalungun Regency in Pematang Raya.Based on data from the Simalungun Regency Central Statistics Agency, the area of wetland rice in Hutabayu Raja District in 2019 as the largest rice production center in Simalungun Regency.Futhermore, a pre-survey was conducted in Hutabayu Raja sub-district to determine the research village.The number of rice farmers in Hutabayu Raja District is 3,636 people.Based on [14] using the formula (equation 1) the sample size was 150 people.

Methods and data analysis
Descriptive statistics and binary logistic regression models were used to examine the data, which SPSS 22 was then used to summarize in tabular form.Descriptive variables were analyzed using means and standard deviations.A binary logistic regression model was used to analyze factors influencing farmers' decisions considering grain quality for GKP sales.
According to [15], the functional form of the logit model is presented as follows: For example, we might be interested in whether or not farmers consider grain quality to sell GKP.The qualitative dependent variable: Di = 1 if ith farmers consider grain quality Di = 0 if ith farmers does not consider grain quality Wether a farmer consider grain quality to sell GKP depend on a number of variables: its age (X1), education (X2), farmers experience (X3), land area (X4) and GKP price (X5).Let us assume that Di is a linear function of variable X, that is, Where ε1 is a disturbance.Assuming E (ε1) = 0, we then have, for given X1… X5, Let Pi = Pr (Di = 1), since Di can only take the value 0 and 1, it follows that Pr (Di = 0) = 1-Pi.Thus: That is, E(Di) is simply the probability that farmer consider grain quality.Rewrite (3): Equation ( 4) is referred to as the linear probability model, could be estimated by OLS but these estimators will not BLUE.The disturbances have a binomial distribution.

𝐸(𝑌 𝑖
Where X1… X5 are explanatory variables.It is necessary to link the unobservable Y* to the observable dummy variables D, by specifying If Y* rises above zero for a farmer that becomes consider grain quality.To obtain an expression for Pi, note that: If a logistic function is used in (6) then we have: The quantity Pi/(1 -Pi) is known as the odds ratio.Notice that in the logit model (8), Pi is not linearly related to X1… X5 as in the linear probality model.

Charateristics of descriptive results
The survey included a sample of 150 farmers with an average age of 53.63 years.The standard deviation is 11.030.The average level of education is 10.95 years, the average experience of a farmer is 22.57 years, the average farm size was 12.962, and the 9.1667 m2 (0.9 ha) as well as GKP price was 4.612 IDR per kg.(table 1)

Determinants of farmers'consideartion towards grain quality to selling grain
The dependent variable encoding show that farmers consider grain quality to sell GKP are classified as 1 while those who will not choose consider grain quality are classified as 0. Based omnibus test of model coefficient used to test the model fit is significant (0.000 < 0.005) hence the model is a good model fit describes the data is a good model.
Based on Hosmer and Lemeshow test show that 0.000 < 0.005 is the sifnificant, it means there is difference between the observed and predicted variable.The Negerkerke square of 0.603 indicates that the predictor variables in the models can account for 60.3% of the change in the criterion variable.Based on the clasification table demonstrate that specifity and sensitivity of the model in relation to predicting group membership on the dependent variable.Grain quality was 94.0 percent, which is anticipated not to be selected based on specificity.The model's anticipated consideration to choose grain quality is 78.8%, with a sensitivity of 78.8%.Overall, the accuracy rate of 90.7 percent was quite high.
Binary logit model results showed that age heads positively and didn't significantly affect farmers' consideration of grain quality at a level less than 5 % significance.The estimated coefficient and odds ratio for this variable was 0.015 and 1.015, respectively.This means that as a farmer's age increases by 1 year, the probability of them considering grain quality increases by 1.015 while other variables remain constant.Older farmers may be used as an experience factor when considering grain quality to sell grain.
This study shows that education level (X2) of farmers negatively and significantly considers grain quality at 5% significance level.based on the theory that one of the factors influencing farmers' decisions to sell good quality grain is better farmer education.The estimated coefficient and odds ratio of the variable are -0.578 and 0.561, respectively.This means that as a farmer's education level increases by one year, farmers' negative propensity to sell quality grain and the probability that they will consider selling quality grain will decrease with probability of 0.561, keeping other variables constant.This may allow farmers with low education levels to consider grain quality for GKP sales.Farmers Experience (X3) in the growing paddy has a significant negative on farmers consideration to sell grain quality.The estimated coefficient and the odds ratio of the variable were -0.117 and 0.889, respectively.As farmers gain experience over the course of one year, farmers in Hutabayu Raja district do not consider the quality of the grain when selling GKP.This means that as farmers in Hutabayu Raja district gain more experience, they do not consider the quality of the grain when selling GKP.Under the assumption of cateris paribus condition, the odds ratio indicates that a unit change of farmer experience on growing paddy decreases the probability of grain quality consideration by 0.889.
Binary logistic model results showed that land area (X4) did not significantly affect farmers' perception of grain quality at less than 5% significance.The estimated coefficient for the the variable was 0.000 and the odds ratio of was 1.000.This means as the land area increases by meter square, the probability of farmers' consideration on grain quality would lead to an increase in their grain quality consideration.The odds ratio indicates that the probability of farmers considering grain quality is equal to the probability of farmers not considering grain quality.It is possible that large farmers are the determining factor in grain quality for grain sales.Under the assumption of the Cites paribus condition, the odds ratio indicates that a unit change in land ownership size increases the probability of the grain quality being considered by 1,000.This result is consistent with related studies [16] showing that the larger the land area, the more water content farmers consider when selling GKP.
This study shows that GKP price (X5) positively and significantly affects farmers' opinions on grain quality at a significance level of less than 1%.The estimated coefficient and odds ratio of the variable are 0.42 and 1.043, respectively.This means that the GKP price increases the probability of farmers being concerned about grain quality by a factor of 1.043, holding other variables constant.It is possible that the increase in GKP price is a factor affecting the quality of seeds sold.Under the assumption of the Cateris paribus condition, the odds ratio shows that a unit change in the GKP price increases the probability of taking into account grain quality by 1.042.This finding is consistent with the conclusion of [17] that farmers selling good quality wheat will receive better prices.Research by [18] shows that based on the president's direction on rice policy, GKP prices tend to increase.and the results of this study [19] show that the implementation of the government purchasing price for dry rice in the shell as directed by the President instruction is effective.This fact shows that the selling price of dry whole rice is higher than the government's purchasing price if it meets the quality requirements of the grain sold.One of the conditions that need to be considered for sustainable agriculture is the existence of price support policies from the Government.

Conclusions
The consideration of farmers towards grain quality when selling GKP depends on various factor such as age, education, farmers experience, land area and GKP price.Additionally, the majority of farmers had a positive outlook towords grain quality, indicating that they asosiated more positive outcomes with considering grain quality when selling GKP.According to this study, a binary logistic regression was used to analyse the impact of different variables on farmers consideration of grain quality.The results show that these variables had a significant effect on farmers consideration.Specifically, the finding revealed that age, land area and GKP price were positively correlated with farmers' consideration grain quality.On other hand, education and farmers experience were negatively correlated with farmer consideration to sell GKP.

Figure 1 .
Figure 1.Map of study area location.

Table 1 .
Summary of samples and descriptive results.

Table 2 .
Binary logistic regression model to predict farmers to consider grain quality.