Re-Tooling the Multiple Choice Test

Multiple choice tests are ubiquitous in colleges and universities, and you’ve all probably taken your fair share, but have you ever wondered where they came from? They didn’t always exist. In fact, they are an invention of the 20^th century. You can blame a guy called Frederick Kelly, who completed his doctorate in 1914 at Kansas State Teacher’s College. He built a standardized reading comprehension test in an effort to stamp out teacher subjectivity when judging work. From there, the MC test exploded into education, aided by World War I, industrialization, and the invention the IMB 8005 Test Scoring Machine. Google it to see a picture of the machine that has brought so many so much stress!! This machine enabled the US military to test 1.7 million potential recruits in a year. Then came the Iowa Test of Basic Skills and the ACT, and the rest is history. A hundred years later, we are still using MC tests to evaluate, rank, and select learners.

High-stakes, timed, multiple-choice tests are normally given at the end of a unit of study to determine if students have "learned" the content. But really, they are a test of:

• Memory (acquisition, retention, and recall/recognition)
• Reading comprehension
• Deductive logic
• Mental endurance
• Calm under pressure
• “What is the exam-writer thinking?”

Perhaps these things are important in your discipline and reflected in your outcomes, in which case, they can be useful. However, MC tests do not effectively test deep conceptual understanding or whether someone is competent. Also, it is very difficult to create good questions. Just look at how long it takes for a data collecting instrument to be validated for research. It requires committee checking, pilot studies, and peer review. Multiple choice questions rarely come under that degree of scrutiny.