January 28, 2014

Teaching for the Test, Part I: Eugenics, Morons, and Mass Intelligence Testing

As a non-credentialed teacher, I encountered a pretty competitive job market after returning to the United States. Not only was it difficult to find a position in education, it was hard to find one anywhere. After countless resumes were sent and cover letters drafted, a combination of desperation and boredom led me to discover a vast and profitable shadow economy: standardized test preparation.

I never took the SAT or ACT, because I had an unusual alternative schooling. However, I did take the GRE in order to apply for graduate school. Since I was used to being overdeveloped in some areas and underdeveloped in others, I decided to study for the exam. Resources were easy to find, and my only purchase was a book from a major company whose sole purpose is providing preparation material for standardized tests.

After a few weeks, approaching the test turned into a game. I was roleplaying, and I upgraded my states by memorizing arcane and useless words, figuring out shortcuts to solving multiple choice math problems, and drilling on essay prompts. I didn't care about the test reflecting my natural intelligence or learned abilities from college. My score would represent the amount of time and effort I committed to cracking the test. I wanted a high score.

The end justifies the means: I took the test and got into a graduate program. I had fallen for one of the greatest (myths) of any standardized exam: to take a test without any preparation is to be buried beneath the scores of everyone else who studies up for it. In an effort to be accepted into Universities, colleges, medical schools, and even private elementary schools, students ironically devote months to memorize arcane vocabulary and adopt techniques impractical outside of exams. In order to learn valuable information, they must first absorb useless data, and pass the test.

The GRE is practically a more adult version of the SAT. I won't reiterate the content of either test here. Both are used as standardized measures of intelligence so colleges and universities can decide whom to include and exclude from their campuses. The SAT claims to be "a reliable, standardized test to assist students, their families, and educators in assessing students' ability to succeed in college-level studies" and is "the most widely used and rigorously researched college admissions test in history."


The origins of standardized testing can be traced to Alfred Binet, who developed the first IQ test, to determine the difference between "normal" and "abnormal" children, the latter of which would be enrolled in special classrooms in the French school system. Robert Yerkes, a Harvard psychologist, adapted the IQ test and gave it to almost two million U.S. recruits during World War I. In a way, Yerkes was bringing the ethos of Fordism to psychological testing: he standardized the test and applied it on a mass scale, with 200,000 tests taken per month. By removing the psychologist from the test - the Binet tests were based on one-on-one exchange between the tested and an experienced facilitator - Yerkes sacrificed empiricism and control for massive results.

Excerpt from Yerke's intelligence test

The conclusion from the tests were modeled on Yerke's eugenicist beliefs: recruits who recently immigrated from Southern Europe under-performed less recent Northern European immigrants, and Yerke used this result to recommend stricter controls on immigration based on national origin. The first large-scale intelligence test in the United States was also the basis for determining racial differences in intelligence: scientific testing could measure innate brain power, not just cultural knowledge, and provide objective proof that certain races or nationalities were less intelligent than others.

Results from Yerke's test

Eugenics and standardized intelligence testing has an interlocking history: Yerke, Lewis Terman, and Henry Goddard were proponents of the belief that genetics defines intelligence, and thus success in later life. Thus, education and social change are less important than discovering who has good genes - by using intelligence tests - and encouraging them to reproduce. Those with bad genes should be discouraged from reproduction. Terman developed the Stanford Binet test with the goal of "curtailing the reproduction of feeble-mindedness and in the elimination of an enormous amount of crime, pauperism, and industrial inefficiency." Goddard, who developed an intelligence test for immigrants to Ellis Island, introduced the term "moron" to the field of intelligence testing, or psychometrics.

While all three claimed to create tests which focused on general knowledge, and minimized testing of cultural facts, this clearly was not the case. In a way, maintaining a cultural bias on these test reinforced the dominant notion - that middle and upper class whites are more intelligent - and extended the argument of racial segregation, immigration control, and eugenics. A critic of Yerke's exam writes:

"Yerkes claimed that the items had been designed to minimize cultural or educational issues, yet some of the questions clearly required familiarity with American culture and history. Subjects had to know that the Overland car was manufactured in Toledo, (not Buffalo, Detroit, or Flint), for example, and that Crisco was a food product and not a disinfectant, toothpaste, or medicine. One sentence-completion item read ‘Washington is to Adams as first is to ____?’"

While the SAT is undoubtedly a product of the history of psychometrics, the College Board, the institute responsible for the exam, predates even Binet's work. The first exam produced by the College Board in 1901, and required students to answer questions in essay form which were rated by experts on a scale from Excellent to Very Poor, with Doubtful in between. An example question from the exam:

1) Write the rules for the following constructions and illustrate each by a Latin sentence :
a. Two uses of the dative. 
b. The cases used to indicate the relations of place.
c. The cases used with verbs of remembering.
d. The hortatory (or jussive) subjunctive.
e. The supine in um.

The SAT was developed in 1926 by Carl Brigham, who worked with Yerkes on the Army intelligence tests. His book A Study of American Intelligence, which he later denounced in 1930, confirmed the native intelligence of Nordic races, and the inferiority of other races:

"Our results showing the marked intellectual inferiority of the negro are corroborated by practically all of the investigators who have used psychological tests on white and negro groups."

The 1926 SAT was a clear move towards objectivity, with a notable section on Artificial Language, where students are given simple grammar rules of a constructed language and asked to translate phrases from English. However, other questions are clearly measures of cultural knowledge:

Indicate which three are thus MOST CLOSELY RELATED:
mutter, pater, sister, descendiente, madre, mère
Bon Ton, Gossard, Djer Kiss, Del Monte, Mavis, Jonteel
Tintex, Globe, Venus, Hammermill, Whiting, Old Hampshire

By the end of the 1930s, the SAT is used as part of the scholarship application for all Ivy League schools. In 1933, IBM machines are used to score tests in New York: thus automation was introduced early on as a grading tool, which perhaps led to the prevalence of the test. Just like Yerke's mass adaptation of the Binet test, machine grading minimized the reliance on experts, and increased the scale and standardization of the exam.

To be continued ...

No comments:

Post a Comment