Simple Bayesian approaches can be applied to answer specific questions in evaluating an individualized reading program. A small reading and study skills program located in the counseling center of a major research university collected and compiled data on student characteristics such as class, number of sessions attended, grade point average, and other demographic characteristics. However, there is no valid way to draw conclusions across such variables. A more meaningful way to present data of this type is to construct a probability tree. Using parametric statistics like means, and standard deviations, correlations require that certain assumptions be met (interval measurement, normal distributions, homogeneity of variance, some variance to begin with, etc.). Standardized reading tests are not adequate criteria of either reading program effectiveness nor do they reflect the reading demands of college courses realistically. Attendance can be a useful criteria for measuring a program's effectiveness. Bayesian technique as applied to decision-making implies that evaluation is a continuous process, and that evaluation is not necessarily concerned with generating new knowledge nor finding ultimate truths which may be the goals of the researcher. Such techniques, used appropriately, can eliminate the expense and effort of gathering of masses of data over a long period of time to make decisions. Arranging demographic and outcome data in Bayesian probability trees makes data easier to understand and interpret. (Contains 11 references and 3 tables of data.) (RS)