Note(s): Table 7 the relation between discrete excess disclosure quality ratings and outcome variables. Discrete excess disclosure quality ratings, INFO_ROUND, are calculated as the difference between raw disclosure quality ratings and rounded predicted disclosure quality ratings based on the model in column (1) of Table 2. The possible values of INFO_ROUND are −3, −2, −1, 0, 1, 2, and 3. The outcome variables include CAR, F_INST, F_ANA, F_FERRORS, F_DISPERSION, F_ENFORCE, and F_ANNLETT. The models are OLS in columns (1) through (5) and Probit in columns (6) and (7). Marginal effects at the mean are reported as coefficients for the Probit models. Detailed variable definitions are in  Appendix 2. t-statistics or z-statistics underneath the coefficients are based on standard errors clustered at the firm. ***, **, and * denote two-tailed statistical significance at the 1%, 5%, and 10% level, respectively

Source(s): Table created by authors