Table 5 presents the results of the regressions set in the equations (4), (5) and (6). The columns (1) and (5) correspond to the equation (4) and columns (2) and (6) correspond to the equation (5) for West Bengal and Rajasthan, respectively .
In both states, the provision of public goods is more aligned on average with women concerns in reserved GPs . This result is interpreted in the following way: "if the difference between the frequency at which a specific request occurs for women and men is 10%, the provision of that good increases by 0.16 standard deviations in West Bengal, and 0.44 standard deviations in Rajasthan" \cite{Duflo2004}. Moreover, the authors also find the average of the difference of this frequency significant, what we called \(\beta_6\) in the equation (5).
The results of the equation number (6) are found in the columns (3) and (7), and the coefficients are not significan. This means that there are not more investment in goods which women (Dij) and both, women and men, (Sij) concern more about in a specific village. Therefore, women are not more responsive or altruistic than men. Then, the differences they found on policy outcomes come from the differences between women and men's preferences. That means, they base their policy decision on their own preferences.
Note that, in columns (4) and (5), they control by the personal characteristics of the Pradhan and the coefficients do not change that much. Also, the authors control in the paper by the PDs reserved for scheduled castes and scheduled tribes and they found the coefficients not significant.