Two inherently contradictory forces are pushing for reform in portfolio assessment. On the one hand there is a felt need for creating more rigid standards that operate to promote uniformity of ratings in appraisal practice to certify achievement. However, on the other hand, critical questions are being raised about separating acclaimed portfolio goals aimed both at appraising achievement while also improving quality of student learning and development. The position of portfolio assessment, which is widely used nowadays in teaching and teacher education, comes into debate. In our study, we look for actual practices in portfolio appraisal in search of criteria used for rating the quality of portfolio materials. It is our interest to find out how appraisal criteria are selected and used to evaluate achievement or to improve the quality of development and learning. In the context of teacher education, we have examined both the espoused criteria of both assessors and collectors of portfolios as well as the actual appraisal practices by looking at the judgmental orientations and supervision styles used by portfolio assessors. In addition, we offered an authentic portfolio document to be rated by different assessors to gauge and compare their quality of rating and criteria use. The actual processes we detected point to a most common practice of employing judgmental, usually normative evaluations based on assessor dependent, more or less pre-decided criteria which permit a "checkbox" approach to appraisal. The findings are mounted up into possible strands for characterizing actual assessment practices of portfolios. In offering a typology on criteria use, transparency could be reached on how to deploy criteria, which could reconcile the teacher educator's dilemma of being both an assessor and a mentor of learning.