3.3. Results from the regression analysis 
3.3.1. Correlation coefficients analysis
Table 3 shows the correlation coefficients between the variables computed using R. Population displays a very strong positive correlation with sorghum production (0.93), strong positive correlation coefficients with the production of maize and rice (0.87 and 0.85), and a moderately positive correlation with precipitation (0.69). Precipitation had a positive strong correlation with rice production (≈ 0.80) and moderate correlation coefficients (≈70) with the production of sorghum and maize. The relationships among the yields of the three crops indicate that sorghum and maize are positively and strongly correlated (0.92), while sorghum and rice (0.82) and maize and rice (0.79) are moderately positively correlated. These results indicate that an increase in both population and precipitation above average values are associated with increases in yields of the three crops. However, population contributed more to the increase in crop production.
 
Table 3. Correlation coefficient values                                
 
3.3.2. Sorghum
 
Using MATLAB, the linear regression model provides a good fit to the Sorghum data with an adjusted R2 of 87.34% and a RMSE of 5.906 x 104 (Figure 4.a). There is a positive correlation between the population and sorghum. After de-trending the sorghum data and fitting the sum of sine and interpolation models (the coefficients are in supplemental III) to the de-trended data, we found several oscillations in the data (Figure 4.b.). These variations are not present in the de-trended data of  precipitation or population.. More data is needed to identify additional key factors in sorghum production.