Factors affecting maize farming’s decision-making process as an attribute in production sustainability analysis in Percut Sei Tuan Subdistrict, Deli Serdang Regency, North Sumatera Province

Maize is a crucial staple food in Indonesia after rice. The decision of farmers to plant maize is driven by many factors which affect their behaviour when deciding to cultivate maize. Deli Serdang Regency is one of the maize-producing centres in North Sumatera Province. Meanwhile, Percut Sei Tuan Subdistrict is one of the subdistricts with the highest planting area, harvest area, and production in Deli Serdang. This research was conducted to determine what factors influence farmers’ decision to cultivate maize, and how the process of taking it occurs. Binary Logistics regression analysis was used to analyse primary data obtained from interviews. The results showed that the farmers’ age, experience in farming maize, number of dependents, and extension attendance frequency, had an insignificant effect on farmers’ decisions to plant maize. In contrast, the ease of farming had a positive and significant impact. Many farmers in the research area cultivated maize since they thought growing maize was less difficult than other crops. The farmers decided to cultivate maize instead of other crops through three stages, namely the knowledge stage, the persuasion stage, and the decision stage. Farmers’ motivation to cultivate maize is one of the attributes in the maize production’s sustainability analysis.


Introduction
Maize is the most important food crop for Indonesian after rice [1].Maize is used as a raw material for various food industries such as vegetable oil, milk, snacks, sweeteners, and flour, as well as the main energy source in animal feed [2].Maize contains various types of nutrients in the form of carbohydrates, proteins, and unsaturated fatty acids, especially linoleic acid [3].Maize oil can be an inexpensive alternative vegetable oil [4].Maize starch content is quite high, making it can be used as a raw material for manufacturing bioethanol by fermentation [5].The stability and continuity of maize production can support various programs in order to realize food and energy security in a state of food and energy crisis [6,7].
In Indonesia, maize can be planted and harvested almost all year round as long as adequate irrigation is available and it gets enough water, where 79.70% of Indonesian maize farmers did not feel a lack of water.So, most maize farmers (88.5%) were not affected by climate change.Hybrid maize varieties 2 were varieties cultivated by the majority (76.87%) of maize farmers, followed by local varieties by 17.08%, and composite varieties were cultivated by the remaining 6.05%.The majority (80%) of Indonesian maize farmers grew maize by monoculture.In addition, 77.92% of Indonesian maize farmers were used to controlling plant pests and diseases chemically, which if done excessively can disrupt the environmental balance [8].
The trend and average productivity of maize in Java tended to be higher than that of other islands.North Sumatera itself was one of the provinces with the highest maize productivity in Indonesia outside of Java.North Sumatera was included in the top 10 provinces with the highest maize production, harvesting areas, and productivity in Indonesia [8][9][10].Figure 1 shows the main maize-producing regencies in North Sumatera.Deli Serdang was the one with the most stable production and harvesting areas of all the regencies.There was no steeply rising or falling trend in Deli Serdang's maize production and harvesting areas, although it was still fluctuating.From 2015-2020, Percut Sei Tuan Subdistrict took turns with Kutalimbaru in the first place as the maize-producing area in Deli Serdang.However, Kutalimbaru's production tended to be more stable, with a drastic increase in 2020, while Percut Sei Tuan was more volatile, and in 2019-2020 it was always lower than Kutalimbaru [11].

Figure 1.
The trend of fluctuations in maize production and harvest areas in the main maize-producing regencies of North Sumatera [12].
Farmers' decisions to cultivate maize are influenced by many factors which ultimately drive farmers' decision-making [13].The farmers' age can influence the farmers' behaviour when deciding to cultivate a type of plant [14].The length of time farmers work on the farm will influence their decision to continue working as farmers [15].Farmers with longer professional experience will find it easier to make decisions for their farms [16].A large number of dependents in the family can motivate farmers to remain involved in agricultural activities to fulfil the daily needs of their family members [17].Farmers' participation frequency in agricultural extension can influence them in the decision-making process regarding their farming [18].Among the attributes that have leverage in influencing the sustainability of maize production, the attributes in the social dimension include the formal education of farmers, the number of farmer households, and farming management patterns [19].Social factors in the maize production's sustainability analysis include the number of farmer households, the formal education, the participation frequency in agricultural extension, the motivation to cultivate maize, the frequency of land use conflicts, the proportion of yield or profit sharing, the participation of family members in farming activities, the consumption per capita, the alternative businesses besides farming, the sustainability perceptions of farming, and the management patterns of farming [20].
Decision-making is usually based on intuition, experience, facts, authority, and rationality [21,22].Intuition is based on emotion which is subjective and easily influenced by psychological factors [23].More experienced people are quicker and more agile in making decisions and solving problems because they are guided by experience [24].Decisions based on facts are more convincing because they are supported by accurate data or information [25].Authority is needed to make decisions on issues that must be resolved.Decisions made must be rational and objective [26].
Age, experience, number of dependents, frequency of attending agricultural extension, and ease of farming are factors that can motivate farmers to plant a crop.The farmers' age will affect the farmers' behaviour when deciding to grow a crop.Older farmers will experience a decrease in physical capacity, memory, courage to take risks, and the level of technological adaptation.The length of time farmers works as a farmer will influence their decision to continue working as a farmer.Farmers who have been in the farming profession longer will be more reluctant to change jobs or the commodities they grow or produce.Farmers with more experience or longer work duration will find it easier and faster to make decisions for their farms.Farming experience also influences their expertise in terms of knowledge and skills in farming so that it supports the success of farming.The more experienced the farmer, the better the performance of his farm.The number of members in the family determines the number of expenses needed by the household.More family members will spend more different expenses.The bigger the family size, the more motivated the farmer will be in working to meet their family members' daily needs.Family members can also become workers on their farms so that farmers do not need to pay for outside family labour.Agricultural extension activities will expand the knowledge, understanding, insight, behaviour, characteristics, and skills of farmers.Farmers who participate more frequently in agricultural extension activities will be more aware of the economic value of their farming activities.The ease of farming will motivate farmers to continue cultivating a crop since the risk of failure is smaller.Farmers' cultivation ability and the type of commodity they cultivate will determine the level of cultivation difficulty.Farmers are more motivated to grow commodities that are more resistant to external risks and less difficult to maintain [27][28][29].
Age-wise, the absence of regeneration due to the lack of interest and motivation of the younger generation to work as maize farmers can reduce maize production.So that Percut Sei Tuan District is able to continue contributing to addressing the challenges of food security and the sustainability of maize farming in North Sumatra, determining what factors influence farmers' decision to cultivate maize, and how the process of taking it occurs are very important.

Methods
Since Percut Sei Tuan is one of the highest maize-producing sub-districts in Deli Serdang Regency where the majority of its farmers grow maize, while its production is still fluctuating and less attractive to the younger generation thus threatening its regeneration and sustainability, this research was 1230 (2023) 012016 IOP Publishing doi:10.1088/1755-1315/1230/1/0120164 purposively conducted there.Since data on the exact number of maize farmers could not be ascertained, based on information from related agencies in Percut Sei Tuan Subdistrict, it was estimated that there were 180 farmers cultivating maize.Of these, the number of respondents who were interviewed as the research sample was calculated using Slovin's Formula [30,31] with an error margin of 10%, so that a value of 64.29 was obtained which was rounded up to 65 respondents.This amount already exceeds the amount of data needed for quantitative statistical tests according to the Central Limit Theorem [32].This number also meets the criteria of 10 times the number of independent variables [33], in which there were 5 independent variables used in this study, namely age, experience, number of dependents, frequency of attending agricultural extension, and ease of farming.Primary data needed for analysis were collected by conducting interviews with sample farmers who were determined randomly using the Simple Random Sampling Method [30,31].
Validity and reliability tests were also carried out on the ease of maize farming variable because it was qualitative data assessed based on scoring so that it is classified as ordinal data [34].Binary logistic regression was used to analyse tabulated interview data because the dependent variable in the analysis was binary qualitative variables (0 and 1) [35], namely the odds of farmers cultivating maize, which is nominal data.For analysis, a random sample of farmers was taken, which was then grouped into 35 farmers who were cultivating maize and 30 farmers who were not cultivating maize (cultivating other crops).Equation 1shows the regression model analysed in this study.

Results and discussion
The statistical descriptions of the independent variables are presented in Table 1.Based on age, it can be seen that all respondents were over 40 years old.This shows the lack of motivation of the younger generation to become farmers in Percut Sei Tuan which can bring a negative impact on food and energy security.Based on experience, farmers were also quite experienced in farming crops.The number of dependents of 5-6 people gave more burdens on the farming household so they have to work harder.However, we need to keep an eye on this since more burdens can reduce the welfare of farmers.A low level of welfare can reduce farmers' concern for the environment and sustainability in running their farms since they want to achieve the maximum production target.Farmers also quite frequently participated in agricultural extension activities.Extension activities can provide farmers with new insight regarding more sustainable and environmentally friendly farming methods.On the ease of maize farming's score, farmers who were cultivating maize scored higher than those who were not cultivating maize.We can see the results of the Binary Logistic Regression analysis in Table 2.The significance values of the step, block, and model in the Omnibus Test resulting from the analysis were less than α 5% (0.000 < 0.05), so H0 was rejected and H1 was accepted.This means that there were significant differences at each step, between block 1 (with independent variables) and block 0 (without independent variables), and between models with independent variables and models without independent variables.Based on that, it could be stated that adding independent variables to the model improved model fit.The significance value of the Hosmer and Lemeshow Test resulting from the analysis was more than α 5% (0.963 > 0.05), so H0 was accepted and H1 was rejected.This means that there were insignificant differences between the data pattern and the binary logistic model.Based on that, it could be stated that the data fit the Binary Regression Model.The Nagelkerke R-Square value of 0.889 shows that the values variation of the independent variables included in the model was able to explain the values variation of the dependent variable by 88.9%.Meanwhile, the rest value of 11.1% was explained by other variables outside the model.All of these results indicate that the Binary Regression Model that had been built was good and fit, so the estimation results were unbiased.The results in Table 1 are then entered into the regression equation in Equations 2 and 3.

Ln
The significance value of the farmers' age was more than α 5% (0.369 > 0.05), then H0 was accepted and H1 was rejected.This means that the farmers' age had an insignificant effect on the farmers' decision to cultivate maize with a coefficient value of 0.084.The odds ratio value of 1.088 (e^0.084 or 2.718^0.084)means that if the farmers' age increased by one year, the odds of the farmers deciding to cultivate maize would rise by 1.088 times compared to farmers who were one year younger.If the other independent variables in the model did not change or remain the same, it means that older people had a higher motivation to cultivate maize (1.088 > 1).
The significance value of the farmers' experience was more than α 5% (0.719 > 0.05), then H0 was accepted and H1 was rejected.This means that the farmers' experience had an insignificant effect on the farmers' decision to cultivate maize with a coefficient value of 0.068.The odds ratio value of 1.071 (e^0.068 or 2.718^0.068)means that if the farmers' experience increased by one year, the odds of the farmers deciding to cultivate maize would rise 1.071 times compared to farmers whose experience was one year lesser.If the other independent variables in the model did not change or remain the same, it means that older people had a higher motivation to cultivate maize (1.071 > 1).
The significance value of the number of dependents was more than α 5% (0.122 > 0.05), then H0 was accepted and H1 was rejected.This means that the number of dependents had an insignificant effect on the farmers' decision to cultivate maize with a coefficient value of 0.898.The odds ratio value of 2.456 (e^0.898 or 2.718^0.898)means that if the number of dependents increased by one person, the odds of the farmers deciding to cultivate maize would rise 2.456 times compared to farmers whose number of dependents was one person lesser.If the other independent variables in the model did not change or remain the same, it means that farmers with larger households had a higher motivation to cultivate maize (2.456 > 1).
The significance value of the frequency of attending agricultural extension was more than α 5% (0.132 > 0.05), then H0 was accepted and H1 was rejected.This means that the frequency of attending agricultural extension had an insignificant effect on the farmers' decision to cultivate maize with a coefficient value of 1.002.The odds ratio value of 2.723 (e^1.002 or 2.718^1.002)means that if the frequency of attending agricultural extension increased by one time/year, the odds of the farmers deciding to cultivate maize would rise 2.723 times compared to farmers whose frequency of attending agricultural extension was one time lesser.If the other independent variables in the model did not change or remain the same, it means that farmers who were more frequently participating in the agricultural extension had a higher motivation to cultivate maize (2.723 > 1).
The significance value of the ease of maize farming was less than α 5% (0.006 < 0.05), then H0 was rejected and H1 was accepted.This means that the ease of maize farming had a significant effect on the farmers' decision to cultivate maize with a coefficient value of 0.652.The odds ratio value of 1.920 (e^0.652 or 2.718^0.652)means that if the ease of maize farming increased by one score, the odds of the farmers deciding to cultivate maize would rise 1.920 times compared to farmers whose ease of maize farming was one score lesser.If the other independent variables in the model did not change or remain the same, it means that farmers who feel cultivating maize tend to be practised easier had a higher motivation to cultivate maize (1.920 > 1).
The decision-making process by farmers consisted of the Knowledge Stage, the Persuasion Stage, and the Decision Stage [36,37].The first stage of decision-making was the stage of knowing something.Farmers already knew of maize cultivation activities in the area where they live.Through interviews, it was known that maize farmers tended to feel that cultivating maize was easier than other crops.The persuasion stage was the stage where farmers had an interest in cultivating or not cultivating maize.A farmer's interest in farming maize could be known through the farmer's knowledge of various kinds of information related to maize, starting from cultivation methods and price information to markets and marketing methods for maize plants.Farmers already had the proper knowledge about maize farming, starting from cultivating maize to harvesting maize properly.Farmers also had good access to price and market information.Marketing that was usually applied was through fellow farmers and maize sellers or selling agents from inside and outside the region.After having an interest in cultivating maize, the next step was for farmers to choose whether or not to cultivate maize.Farmers chose to cultivate maize by taking into account the positive things and benefits that would be obtained from maize farming.If the farmer felt that he had the right reasons, the farmer would decide to cultivate maize.

Conclusions
To summarize everything in this study, we could see that the farmers' age, experience in farming maize, number of dependents, and extension attendance frequency, had an insignificant effect on farmers' decisions to plant maize.In contrast, the ease of farming had a positive and significant impact.Many farmers in the research area cultivated maize since they thought growing maize was less difficult than other crops.The farmers decided to cultivate maize instead of other crops through three stages, namely the knowledge stage, the persuasion stage, and the decision stage.Farmers' motivation to cultivate maize is one of the attributes in the maize production's sustainability analysis.
= b 0 + b 1 Fa + b 2 Fe + b 3 Fm + b 4 Ae + b 5 Ef (farmers cultivating maize (1 = cultivating maize, 0 = not cultivating maize) Fa = farmers' age (year) Fe = farmers' experience (year) Fm = number of dependents (people) Ae = frequency of attending agricultural extension (times/year) Ef = ease of maize farming (score) b 0 = constant/intercept coefficient b 1 , b 2 , b 3 , b 4 , b 5 = parameter/regression coefficients Parameter estimation was carried out with the utilization of SPSS 24 software.The goodness of fit model was assessed through the Omnibus Test, Hosmer and Lemeshow Test, and the Nagelkerke R Square value.Then the Wald Test was used to see the significance of the independent variables' effect on the dependent variable partially at the 95% Confidence Interval.

Table 1 .
The statistical descriptions of the independent variables.

Table 2 .
The results of the Binary Logistic Regression analysis.