Multiple Linear Regression Model for Analyzing the Determinants of Rice Production in Sumatra

Rice production was a crucial component in maintaining national food security, particularly in Sumatra, which was known as one of Indonesia's rice granaries. This study aimed to analyze the determinant factors affecting rice production in Sumatra using a multiple linear regression approach. Secondary data utilized in this research included 224 observations from eight provinces in Sumatra, covering the period from 1993 to 2020. The independent variables analyzed included Year, Harvest Area, Rainfall, Humidity, and Average Temperature, with Rice Production as the dependent variable. The resulting multiple linear regression model was expressed by the following equation: Rice Production (Y) = −63.442.498,904 + 32.535,572 × Year + 4,622 × Harvest Area + 46,106 × Rainfall – 3.993,403 × Humidity – 62.701,579 × Average Temperature. Data processing was performed using SPSS software, which included normality testing, linearity testing, simultaneous F-test, and coefficient of determination analysis. The normality test results indicated that the residuals of the model were normally distributed (Asymp. Sig. = 0.227 > 0.05). The linearity test showed a significant linear relationship between the independent variables and rice production, with no deviation from linearity. The regression model demonstrated a coefficient of determination (R²) value of 0.877, meaning that 87.7% of the variation in rice production could be explained simultaneously by these five variables. Furthermore, the F-test results indicated that the regression model was statistically significant (Sig. = 0.000). These findings suggested that the five variables significantly influenced rice production and could serve as a basis for strategic decision-making in the agricultural sector. The resulting model also held the potential as a predictive tool in planning policies to enhance rice production in the future.