Students answered only 35.5% of the questions correctly on the first one.
Before 1845, American schools measured learning through oral recitation. A teacher would call on a student, ask questions, and judge the answers by ear. The method favored children with confidence, social fluency, and fluent English, and it left no permanent record.1
Horace Mann, secretary of the Massachusetts State Board of Education, believed written exams would be more objective. After visiting Prussian schools in 1843, he returned convinced that a printed test could produce what he called a "daguerreotype likeness" of each student's mind, a record that could be carried away and compared across schools.2
In 1845, Mann had examiners prepare and administer written tests to students in Boston's grammar schools. The teachers had not seen the questions in advance. Five hundred and thirty students, selected as the best below high school age, sat for one-hour written exams in geography, grammar, history, rhetoric, and philosophy.3
On average, students answered only 35.5% of the questions correctly.4 Mann released the results to the public and used them to criticize the quality of teaching in Boston. Teachers responded that the exam questions had little connection to what they had actually taught.
The resulting conflict led to teachers being fired and school board members losing their seats. The effectiveness of the method itself was not questioned. As one congressional report later noted, it was taken as an article of faith that test-based information could improve a bureaucratizing school system.5
The Army Alpha and Army Beta tests, developed during World War I to sort 1.7 million recruits by cognitive ability, became a model for civilian schools in the 1920s.6 The SAT followed in 1926, originally designed by Carl Brigham, a Princeton psychologist who had worked on the Army tests. By the time the No Child Left Behind Act passed in 2001, federally mandated standardized testing reached every public school in the country.7