1219 PM Wed Nov 11 6 Problem Set 2_2.pdf Question 3 (10 points) 3 of 3 This question asks you to make sense of regression results based on the information provided. In the British colonial empire, the Governor of a colony was a bureaucrat appointed by London. Before 193 0, governors were appointed in a patronage system by the Secretary of State (a politician). Aer 1930, they had to go through a civil service exam. In the table below, each column is from a separate regression. The dependent variable is a governor's salary in a given year, in logs. The first row shows the mean of the dependent variable. The next 2 sets of rows are coefcient estimates for the listed variable with standard errors given in parentheses. \"Connected\" is equal to 1 if the governor in ofce is socially connected to the Secretary of State (e.g., is a relative or a former classmate), and 0 otherwise. \"Reform dummy\" is 1 aer the 1930 reform and 0 before. [Note that the table shows a lot of things which you won't need.] Describe what we learn from this table and briey discuss. [1-2 paragraphs] Governor salary (l) (2) (3) (4) (5) (6) Mean of dep. var 7.929 7.929 7.929 7.929 7.929 7.986 Connected 0.097 0.127 0.205 0. l29 0,091 0,] 72 (0.036) (0,043) (0.059) (0.043) (0.041) (0.045) Reform dummy x connected 0.l23 0.222 0.119 0.l38 0.205 (0.056) (0.079) (0.058) (0.078) (0.056) Connected + mform dummy 0.004 0.017 0.009 0.046 0.033 x connected (0.040) (004]) (0.048) (0.068) (0038) Year FEs Yes Yes Yes Yes Yes Yes Governor FEs Yes Yes Yes Yes Yes Yes Spell length FEs Yes Yes Yes Yes Yes Yes Time-varying controls Yes Yes Yes Yes Yes Yes Connected x trend (centered I930) Yes Connected x governor characteristics w v v Yes ~ ~ Connected x colony characteristics Yes Sample Full sample Pre/ post Observations 3.510 3,510 3,510 3.510 3,510 2.429