http://hastac.org/files/imagecache/homepage_50/pictures/picture-79-873560aec16bee4b69793f2fa0fbd715.jpgBy Cathy Davidson. In 2009, when I was researching and writing the "How We Measure" chapter of Now You See It, I could offer a critique of the current system of standardized, summative, end-of-grade testing, but had little to offer that addressed the key positive contributions of standardized testing:  its ability to be graded by machine (or by humans with a template) and provide the basis for comparisons of outcomes across disparate institutions. Originally pioneered in a doctoral dissertation at Kansas State Teacher's College by Frederick Kelly in 1914, the item-response test was adopted widely and almost immediately because of these two features.
Many esteemed critics over the years have pointed out that multiple-choice tests are a poor instrument for motivating learning, for  instilling an appreciation for complexity, or for accurately testing what one actually learned. They encourage “teaching to the test” and “learning to the test,” especially in our post-2002 No Child Left Behind world where public school teachers can be penalized and schools closed if kids don’t do well on such tests.  However, until another system addresses the key attributes of standardization and automation, school life will continue to be governed, preschool to professional school, by EOGs, SATs, ACTs, GREs, LSATs, M-CATs and the like. Read more...