The regression estimates for different locational and farm-specific variables are given in Table 7. Many of the variables have the expected signs and are significant at 5 per cent or lower level. For example, the Old Alluvial Region (OAR) has the highest average level of (group) TE, exceeding the corresponding measure for overall by 0.139 (Table 4) and has lowest variability in the levels of efficiency. In the regression analysis reported in Table 7, the coefficient of OAR dummy variable is only 0.046. Moreover, it is not even statistically significant. By contrast, such difference for the Red and Laterite Zone (RLZ) is 0.079 in Table 4 and the estimated coefficient of the RLZ dummy variable in Table 7 is comparable being at 0.115, and significant at one per cent level. This shows that controlling other factors sometimes (though not always) could portray a different picture about technological differences across the regions.