Abstract
This chapter describes Coker College’s subjective performance assessment program to rate student thinking and communication skills. It uses a discussion of the epistemology of assessment to motivate an emphasis on direct observation by experts as the basis for “authentic” assessment for complex learning outcomes. Results from Coker College’s experience are given and discussed in the context of this approach. The purpose of this chapter is to give a philosophical framework and practical methods that can help institutions assess liberal arts learning outcomes. Such assessments can provide information crucial to improving programs and pedagogy and form the basis for institutional effectiveness reports to stakeholders.
TopIntroduction
Because this is a case study, we will be interested in implementations and details and uses of data. But these are more valuable if we know the rationale behind the assessment methods, so we begin with an examination of the measurement of learning outcomes. The questions we address are in turn theoretical and practical: What can we know, and how can we know it? How does work in practice? Answers to these questions are explored in detail in the following sections:
• The Nature of Assessment
• Measurement and Reality
• Measurement and Probability
• Subjective Measurement
• Authentic Assessment in Practice
• Defining Outcomes
• Gathering Data
• Institutional uses of Assessment Data
• Program uses of Assessment Data
• FACS in the Classroom
• Reliability and Validity
• CONCLUSION
• Advantages
• Challenges
• Conclusion
It can be the case that an assessment program is put into place without asking these questions, but this may invite confusion when the results arrive.
There are many options open to an institution wishing to assess general education or liberal arts skills like thinking and communication. Standardized testing and portfolio assessment are two examples at opposite ends of a spectrum of possibilities. In this chapter we will derive a kind of performance testing.
Performance testing includes such varied procedures as observing a foreign-language student having a conversation in the foreign language, requiring science students to conduct a real experiment, asking students to work together as a group and observing the interaction, and giving problems that have no answer or more than one correct answer and observing a student’s approach. (Kaplan & Saccuzzo, 2004, p. 631)
The description above makes it clear that the observer is crucial to this form of assessment. At Coker College, course instructors use normal class work to find opportunities for observing performance. As we shall see, there is a trade-off between this hands-on approach and objectivity (and hence reliability), but the results will support this decision.
We will first seek an understanding of what measurement is, and in the process reject overly deterministic models like fill-in-the-bubble tests for the purpose of assessing complex outcomes. We are not alone in this: “The ‘new’ assessment currently being called for by the National Education Association and others is performance testing.” (Kaplan & Saccuzzo, 2004, p. 631).
True performance testing tries to maximize reliability by carefully controlling testing conditions. See (Harris, 2002, pg. 81) for a list of testing components including “[…] specification of the steps or actions required to perform the test, typically in the order in which they are to be performed.” This will prove to be a bit too restrictive for our agenda of assessing educational outcomes that are not the results of a step-by-step process. We will use the term “authentic assessment” instead, although in the literature the two terms are sometimes used interchangeably. In our case “authentic assessment” will depend less on rigid specifications and more on subjective judgment by experts.
Key Terms in this Chapter
Core Skill: At Coker College, the core skills taken from the institutional mission are effective writing, effective speaking, analytical thinking, and creative thinking.
Complexity: The minimum length for a complete description of a process. Also called Kolmogorov complexity. In educational assessment more complex phenomena manifest themselves in more various forms.
Virtual: Existing only in the informational sense, as a combinatoric property of matter. A chess position is virtual; the arrangement of the pieces has importance, not their physical properties.
FACS: Faculty Assessment of Core Skills. A process of routinely gathering authentic assessments of core skills, based on observations taken during coursework.
Liberal Arts: Liberal arts institutions profess to place special value on general thinking and communication skills, and a broader view of education than technical or vocational training.
Authentic Assessment: Sometimes defined to be synonymous with performance testing. Here we use it explicitly to mean a subjective judgment resulting from direct observation of performance by an expert.
Event: Something that can happen with some probability. Theoretically, events belong to some sample space of possibilities. Different observers may judge events differently.
Performance Testing: Evaluating performance based on observation of the subject performing the task under controlled conditions.
Measurement: Producing a numerical value with a meaning most observers will agree with. Measurements are usually in units that aggregate, but may be virtual or real.