Is the result sensitive to the choice of scores? Re-fit the linear probability model using scores 0, 1, 2, 3, 4, and re-evaluate fitted probabilities at alcohol levels 0 and 7 and the relative risk. d. - egory than the others because, although it has only one malformation, its sample size is only 38. Are results sensitive to this single malformation? Fit the logis- tic regression or linear probability model with and without that observation, and evaluate fitted probabilities at alcohol levels 0 and 7 and the relative risk.