| The current concept of assessment came to the forefront through national attention on quality and accountability in higher education. In its recency, assessment of quality in higher education has provided practitioners at institutions few tools to guide the assessment process. Sets of general assessment principles and standards that could provide valuable assistance are missing from the literature. Assessment standards found in the literature were inconsistent, incomplete, and widely dispersed. The goal of this research was to produce a set of standards agreed upon by assessment experts as the most important in the conduct of quality assessment in higher education.;The Delphi, a group technique to produce consensus, was used with a group of 85 higher education assessment experts. An inventory of 40 standards was compiled from existing sets of evaluation standards of the Joint Committee on Standards for Educational Evaluation and of the Evaluation Research Society. A three-round Delphi identified the most important standards for each of the five phases of assessment defined in the research: Clarifying Object and Context, Designing the Assessment, Collecting and Analyzing Data, Communicating Findings, and Using Findings. In addition to rating the importance of the standards in two rounds, the participants provided rationale statements for the standards they considered most critical for each phase. The number of experts participating in the three rounds were 70, 53 and 38 respectively.;On the basis of the final ratings, a set of 16 standards was produced. Twelve of the standards were identified as very important in at least one phase of assessment. Four additional standards consistently played an important role across the phases of assessment and were rated as important to some degree in at least four of the five phases. None of the standards were identified as very important in Phase V. One standard was not identified as being important in any of the phases.;Few importance ratings, about 5%, were changed in the second round re-rating. No significant relationships were found between the number of changes made by participants and several background variables. The importance assigned to some standards was significantly related to various background variables of the participants. |
