When we reject the null hypothesis of a partial f-test, then we conclude that a. at least one of the eliminated variables is significant. therefore we should use the reduced model. b. at least one of the eliminated variables is significant. therefore we should use the full model. c. the eliminated variables are jointly not significant. therefore we should use the reduced model. d. the eliminated variables are all significant. therefore we should use the full model.