Advisor(s) - Committee Chair
Reagan Brown (Director), Katrina Burch, J. Farley Norman
Department of Psychological Sciences
Master of Science
Department of Psychological Sciences Western Kentucky University There are two options to estimate a range of likely values for the population mean of a continuous variable: one for when the population standard deviation is known and another for when the population standard deviation is unknown. There are seven proposed equations to calculate the confidence interval for the population mean of a dichotomous variable: normal approximation interval, Wilson interval, Jeffreys interval, Clopper-Pearson, Agresti-Coull, arcsine transformation, and logit transformation. In this study, I compared the percent effectiveness of each equation using a Monte Carlo analysis and the interval range over a range of population means to determine the accuracy of the equations. Results indicated that the Agresti-Coull equation and Clopper-Pearson equation are the most successful at locating the population proportion at least 95% of the time across the range of population proportions and that the Agresti-Coull equation has the narrower interval range.
Applied Statistics | Industrial and Organizational Psychology | Statistical Models
DuBose, Morgan Juanita, "A Monte Carlo Analysis of Seven Dichotomous Variable Confidence Interval Equations" (2022). Masters Theses & Specialist Projects. Paper 3568.