An analytical technique from the field of market research called conjoint analysis was applied to a psychological measurement of student testing design preferences. Past concerns with testing design are reviewed, and a newer approach to testing is identified--the modified confidence weighted-admissible probability measurement (MCW-APM) test scoring method developed by J. Bruno (1985). The conjoint analysis approach was applied to measure student preferences for testing design. The conjoint design was developed with four factors: objectivity (subjective and objective), restrictions (restricted and unrestricted), partial credit (allowed and not allowed), and feedback (given and not given). Usable survey responses were received from 516 individuals from different levels of education and different colleges. To demonstrate the flexibility of conjoint analysis, the study design was modified to add a fifth factor of test-responding simplicity. The illustration indicates that conjoint analysis can be used for larger sample sizes and that it permits analysis by demographic factors, a feature that is applicable to education. Six tables present study data. A 12-item list of references is included. (SLD)