A Glossary of Assessment Terms
In order to develop this guide, it was necessary to define and explain terms as they seemed appropriate for the purposes of this guide. ACAA recognizes that some terms are understood in ways different from that used here, concepts such as assessment, methods, measures, goals, objectives, and values to cite a few. In an effort to be helpful, not arbitrary, the following explanations of terms as used in this guide are offered.
Approaches are the procedures used to gather the information needed to assess how well students have met the learning objectives. They are the course of action through which evidence about courses, programs, majors and the like will be gathered. To provide quality information, multiple approaches should be used.
Assessment refers to a continuous process instituted to understand and improve student learning. While academic units may find alternative pathways to arrive at this goal, this process needs to begin with articulation of educational goals for all programs and courses. These goals should be expressed as measurable objectives followed by the selection of reliable and valid methods and measures. After collecting, interpreting, and sharing findings, the aim is to use these learning outcomes to better understand how and what students learn, how well students are meeting expected objectives, as well as to develop strategies to improve the teaching and learning processes.
Benchmark is the actual measurement of group performance against an established standard or performance, often external.
Criterion is the standard of performance established as the passing score for a performance or other measures such as a test. The performance is compared to an expected level of mastery in an area rather than to other students’ scores.
Cross-Sectional Studies provide information about a group of students at one point in time.
Evaluate and Evaluation are terms used in this guide to indicate the interpreting of findings and are used as synonymous to the term assess and assessment. ACAA is aware that many make a distinction between evaluation and assessment with the difference that assessment is a process predicated on knowledge of intended goals or objectives while, in contrast, evaluation is a process concerned with outcomes without prior concern or knowledge about goals. That distinction is not used in the guide.
Goals are statements about the general academic aims or ideals to which an educational unit aspires. Goal statements allow us to share with others our hopes in regard to the learning achievements of our students. Further, goals at the unit level should align with the mission of the university. Goal statements are not amenable, as stated, to measurement.
Longitudinal studies provide information from the same group of students at several different points in time.
Measures are the specific instruments or performances used to provide data about learning. They are the tools that are to provide information as to the level of achieved results or outcomes. To avoid systematic bias in findings, multiple measures are required.
Methods - see approaches.
Objectives are the redefinition of learning goals in a way that permits their measurement. Objectives express the intended results or outcomes of student learning and clearly specify the criteria by which student knowledge, performance, or values will be evaluated.
Process is a method generally involving steps or operations that are ordered and/or interdependent.
Qualitative and Quantitative Research describe two research methods. Both are valuable as a means to assess student learning outcomes. In a practical and somewhat philosophical sense the difference is that quantitative research tries to make use of objective measures to test hypotheses and to allow for controlling and predicting learning. Qualitative research makes use of more subjective observations of learning.
Reliability is the extent to which studies or findings can be replicated.
Sampling consists of obtaining information from a portion of a larger group or population. When the selection of a sample is randomly chosen there is greater likelihood that the findings from the sample will be representative of the larger group.
Validity depends on demonstrating that a measure actually measures what it is purported to measure.