(1) Edited table 4 top match the style of other summary tables.
(2) Discovered why the observation numbers were different between old and new tables: the old table described the full data while the new one partially describes the full data but focuses on the data used in the regressions.
NOTE: The review sentiment averages seem very similar. This is not a coding mistake. They are simply very similar between the full and reduced dta sets.
(3) The previous merge branch commit was a minor merge error which I fixed on my end.
(1) Edited table 4 top match the style of other summary tables. (2) Discovered why the observation numbers were different between old and new tables: the old table described the full data while the new one partially describes the full data but focuses on the data used in the regressions. NOTE: The review sentiment averages seem very similar. This is not a coding mistake. They are simply very similar between the full and reduced dta sets. (3) The previous merge branch commit was a minor merge error which I fixed on my end.