All estimates and further details about each model specification are reported in Table A3 in S1 Online appendix. Standard errors are estimated allowing for clustering at the state-year level. Horizontal bars around point estimates illustrate 95% confidence intervals. The following controls are added sequentially in the other models: electoral record and political controls (Model 2), economic controls x majority party membership (Model 3), state fixed effects (Model 4), and state fixed effects x majority party membership (Model 5). Each row corresponds to a different model specification: Model 1 includes only time dummies and dummies for majority party membership and membership of the governor’s party as controls. Panel (a) shows results for outcome 1 (re-election), while Panels (b) and (c) show results for outcomes 2 (replaced by same-party candidate) and 3 (replaced by other-party candidate), respectively. The figure shows average marginal effects of changing the budget gridlock variable from zero to one on the probability of each of three possible election outcomes for incumbent legislators, allowing different effects for members of the governor’s party vs. All estimates and further details about each model specification are reported in Table A2 in S1 Online appendix.Įffects of budget gridlock: members of governor’s party vs. The figure shows average marginal effects of changing the budget gridlock variable from zero to one on the probability of each of three possible election outcomes for incumbent legislators, allowing different effects for majority vs.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |