A standard-setting study is a formal research study, conducted by an organization that administers a test, used to determine the cutscore, or passing threshold, for that test. In the United States, a legally defensible cutscore for a high-stakes assessment cannot be set arbitrarily and must instead be empirically justified to meet the Standards for Educational and Psychological Testing; an organization cannot simply decide, for example, that the cutscore will be seventy percent correct. Instead, a study is conducted to determine what score actually distinguishes between classifications of test-takers, such as competent versus incompetent. Because such studies require substantial resources and a number of professionals with a psychometric background, they are impractical for an ordinary classroom, even though some form of standard setting takes place at every level of education using a range of available methods. A standard-setting study is typically carried out using a focus group of five to fifteen subject-matter experts who represent the test's key stakeholders, such as instructors familiar with the capabilities of the population being tested.
Facts
Core ClaimA legally defensible cutscore separating passing from failing performance on a test cannot be set arbitrarily but must be empirically justified through a standard-setting study, typically a panel of five to fifteen subject-matter experts applying either an item-centered method that rates each item's difficulty for a borderline candidate or a person-centered method that compares the score distributions of examinees already sorted into performance categories. 1 Sources
1. Standard-setting study (Wikipedia)
Wikipedia- lead section
Item-centered studies footnote, Nedelsky entry
Nedelsky, L. (1954). Absolute grading standards for objective tests. Educational and Psychological Measurement, 14, 3-19.
View the Source Reader Challenges (0)
No disputes yet. Spotted an error or a better source? Open the first one.
Sign in to dispute this or suggest a correction.