Our information-oriented society relies on the widespread use of color CRT displays. Misconvergence of the primary colors of a shadow-mask CRT is a problem with this technology that deserves human factors engineering consideration. The purpose of this research was as follows: (1) to determine the effects of misconvergence type and amount and font type on reading performance (time required and errors made), (2) to determine the effects of misconvergence type and amount and font type on subjective image quality ratings, (3) to determine the role of luminance and chrominance contrasts in predicting performance or subjective ratings. Ten participants performed a simple reading task and rated the image quality of the text they had just seen using a nine-point scale. The text was presented on a shadow-mask CRT. Different misconvergence types and amounts and different font types were presented. Neither font type, misconvergence type, nor misconvergence amount affected the time required to perform the reading task. Only misconvergence type affected the rate at which errors occurred, with blue misconvergence of a white character resulting in the most errors and cyan misconvergence resulting in the fewest errors. Font type, misconvergence type, misconvergence amount, and the misconvergence type and amount interaction all affected subjective ratings, with 1 to 2 arcmin being the largest acceptable misconvergence amounts. The Yu'v’ chrominance contrast between the stationary misconvergence fringe and the background was positively correlated with subjective preference ratings. / Master of Science