The History of the Multiple-Choice Question

Multiple-choice standardized testing is a ubiquitous feature of modern life. We encounter questions with four or five options and only one right answer almost everywhere – from Mrs. Anderson’s fifth-grade history class to driver’s permit tests to the game show Who Wants to be a Millionaire. It might come as a surprise, then, to learn that the multiple-choice question did not even exist in the 19th century. As a way of finding out what someone knows, the idea of a question with a fixed number of choices and a high degree of supposed objectivity was invented in the early 20th century, a product of the era’s instinct for mass production. The same forces that gave us the Model-T also gave us “Which of the following best completes the sentence?” And the multiple-choice question became an essential tool for the new educational theories of the industrial age.

The father of multiple choice testing was one Frederick J. Kelly, author of the Kansas Silent Reading Test. In 1914, Kelly devised a system that he believed would take subjectivity out of testing and create an efficient means of gauging students’ reading ability. The multiple-choice question was crucial to accomplishing these aims. (Ironically, Kelly would later regret the impact of standardized testing and attempt to reverse course as the president of the University of Idaho, only to be dismissed from his position for resisting ‘modernization.’)

The instructions for administering the Kansas Silent Reading Test looked like this:

The similarity between this original standardized test and all that would follow is striking. Anyone who has taken a standardized test (and who hasn’t?) can recognize the basic pattern:

• “Do not open the papers until you are instructed to do so.”
• “Work fast but smart.”
• “Stop at once when time is called.”
• “Do not use the wrong marking process or your answer will not count.”

The only real difference between this and the SAT is that the Kansas Silent Reading test was a five-minute exercise while the SAT demands over three hours.

In 1937, standardized testing got a further boost from an invention by Reynold Johnson. Johnson (the “father of the hard drive”) was the first to create an instrument for scoring tests electronically. Originally designed for his own use as a teacher, a production version of the machine would later be released as the IBM 805 Test Scoring Machine. The effectiveness of machine scoring was then improved upon in the 1950s and ’60s through Everett Franklin Lindquist’s pioneering work in optical mark recognition. (Lindquist was the architect of the ACT, the GED, and NMSQT.) Finally, millions of students could be efficiently processed by the education-industrial complex using assembly line procedures – an effort that found its ultimate trajectory in instruments like the Iowa Basic Skills Tests, the SAT, and the ACT.

There is no doubt this massive standardization of testing has made it easier and more efficient to administer tests, gather data, and sort students. In many ways, it is an impressive achievement. However, in the face of all this standardization and efficiency, we need to begin asking: “To what end and with what results?”