The validity of assumptions that portfolios complement other assessment methods and yield more reliable and valid data than do traditional methods was studied. More specifically, focus was on examining whether: achievement level (characterized by scores, ranks, and group membership) varies significantly with differing definitions and usage of portfolios; and progress determined through portfolio assessment differs from progress ascertained using a variety of traditional assessment methods (such as paper and pencil tests, interviews, and cumulative grade point average). Data on 30 college students in teacher education enrolled in three schools of the Alabama Consortium for Minority Teacher Education were collected during the freshman year and at the end of the sophomore year via standardized tests, questionnaires, and portfolios. Portfolios included students' exhibits of milestones reached, activities that they considered enriching to personal, social, and academic development. Data were analyzed using the Pearson correlation, multiple regression, and discriminant analysis techniques. Results show that performance measured using the portfolio technique is related to performance determined using traditional measures such as ACT (American College Test), college grade point average, and essays. However, the nature of that relationship suggests using portfolios as a supplement, rather than as a substitute, to traditional measures. Two tables present study data, and a 28-item list of references is included. (SLD)