A (Mostly) Brief History Of The SAT And ACT Tests

late 1800s

Average Scores

SAT (1952–present)
Data | Graph

ACT (1970–present)
Data | Graph

By the end of the 19th century, tests for admission to U.S. colleges are specific to each school. (The arithmetic portion of Harvard's 1869 entrance exam can be seen here. The corresponding portion of MIT's 1869 exam can be seen here. These portions of the two tests are quite different in difficulty.) The content of the tests varies widely and can be highly dependent on the interests of the faculty conducting the exams. It is not unusual for a college to administer exams on campus a week or two before classes begin. As an alternative to testing, many colleges, especially in the midwestern U.S., use "admission by certificate": a high school would be certified by inspectors from the colleges to have an appropriately preparatory curriculum for college work, and students graduated from such a high school would be considered to be adequate without testing. By 1900, however, inspections are infrequent and college faculty are often not present.

1900

The College Entrance Examination Board (or "College Board") is founded, consisting of a non-profit membership of twelve colleges and universities. The membership is comprised mostly of elite institutions in the northeastern U.S., including Columbia, Cornell, Vassar, Barnard, and New York University. The founders are worried that the lack of uniform admissions testing and the certificate system places too much control of college admissions with the high schools. Also, the certificate system is thought to be conducing students away from the northeastern colleges.

At this time, roughly 4% of high school graduates go on to college.

1901

The "College Boards" are administered in June for the first time to fewer than 1000 students. Roughly 75% of these students are applicants to Columbia University, hence the practical effect of these tests is to distinguish excellent students from elite students. The essay tests are curriculum-based achievement exams, designed to assess a student's mastery of nine subjects, including Greek, Latin, and physics. Scoring is done by hand and consists of five ratings for each subject, from "Excellent" to "Very Poor".

1926

The Scholastic Aptitude Test (or "SAT") is administered for the first time to about 8000 students, 40% of whom are women. (Almost all of these students are taking the traditional boards as well.) Carl Brigham, a psychologist who helped to develop aptitude tests for the U.S. Army during World War I, is influential in the development of the 1926 test. The SAT is considered a "new psychological test" and a supplement to, but not a replacement of, the existing College Boards. Due to the completely different nature of the SAT compared to the boards, all students are required to take a practice test before the actual SAT (sample questions below). The majority of the content of the first SAT is taken directly from Brigham's "Princeton Psychological Examination", which itself was derived from the Army "Alpha" intelligence tests.

Unlike the College Boards, the three-hour SAT (administered in June) is designed primarily to assess aptitude for learning rather than mastery of subjects already learned. For some college officials, an aptitude test, which is presumed to measure intelligence, is appealing since at this time intelligence and ethnic origin are thought to be connected, and therefore the results of such a test could be used to limit the admissions of particularly undesirable ethnicities. The test is designed to assess ability independently of any particular secondary school curriculum, which has a more mainstream appeal: college admissions testing via the SAT is uniformly applicable across a wide range of high school students, and the test is firmly in the control of college officials.

The 1926 SAT is very hard for most students to finish: 315 free-response questions in 97 minutes. The instructions include the following: The pencil is preferable to the fountain pen for use in this sort of test. The test is comprised of nine sub-tests: two math tests (Arithmetical Problems, and Number Series), and seven verbal tests (Definitions, Classification, Artificial Language, Antonyms, Analogies, Logical Inference, and Paragraph Reading). Raw scores on the sub-tests are converted to a single scaled SAT score ranging from 200 to 800, with a mean of exactly 500. Using this scoring method means that an unusually strong group of students taking the test could push other students' scores down, unlike the modern SAT. Also, the scores in a particular year could not be compared with scores in another year. For example, a student obtaining a score of 600 in 1926 could be significantly weaker than a student obtaining a score of 600 in 1927, if the group of test takers in 1927 happened to be particularly good students overall compared to 1926.

The questions in the reading sections include six-choice antonyms, analogies, and artificial language translations. A practice test given to students taking the 1926 test includes the following six-choice antonym question (there are six possible pairs of numbers as answers):

Which two of the following four words are opposite or nearly the opposite:
1) obedient; 2) sincere; 3) dissembling; 4) torpid.

An example of a "classification" question is below (there are twenty possible answers):

Which three of the following words are most closely related?
1) bean; 2) potato; 3) carrot; 4) beet; 5) lettuce; 6) cabbage.

A typical "number series" math question on the 1926 SAT asks the student to complete the sequence given by filling in two numbers at the end. A difficult example from the 1926 practice test:

Which two numbers come next in the sequence: 1/8, 1/8, 1/4, 3/4, 3, ?, ?

Other math questions are open-ended arithmetic word problems, such as the following:

A boat that can make forty miles an hour in still water makes a trip of one hundred miles down a certain stream. If this trip takes two hours, how long will the return trip take?

(Answers appear at the end of this timeline. Remember, you have about 20 seconds per question.)

As with the modern SAT, the original 1926 SAT and successive tests have an "experimental" section (30 minutes in length) which is used to test new questions and question types. The section is not identified to the test takers, requiring them to apply themselves fully to this section as well.

1928

The Artificial Language and Logical Inference sections are dropped from this year's SAT (never to appear again). Both math sections are removed from the test as well.

1930

Free-response math questions reappear for the 1930 test. Test takers are expected to solve 80 math questions in 100 minutes. The SAT is split into a "verbal aptitude" section and a "math aptitude" section, and a score on a 200 to 800 scale is reported for each of these sections. These scores are not sent to either the student or to his or her high school: only colleges and universities receive scores at this time.

1933

Harvard begins using the SAT in order to determine the quality of applicants for a new scholarship program. The university feels that the test provides an accurate assessment of a prospective student's intelligence. (The scholarship was designed to attract students other than those from well-known East Coast private schools.) By 1938, all of the College Board member schools will be using the SAT to evaluate scholarship applicants.

1936

Math is once again removed from the SAT.

1937

The College Board's Achievement Tests are administered for the first time in April; each test is a multiple-choice format assessment of proficiency in single subjects such as biology, chemistry, and math, among others. The Achievement Tests are scheduled in the afternoons of the SAT test days, which are now also offered in April as well as June.

Secondary schools are given the SAT scores of their students for the first time starting in this year; whether or not students can learn their own test scores is up to the high school.

1940

After a slow growth in acceptance of the SAT during the 1930s, the number of test takers exceeds 10,000 for the first time in April. (The total number of U.S. high school graduates in 1940 is roughly 1.1 million, meaning that only about 1% of these graduates take the SAT.)

1941

The verbal portion of the SAT in this year is curved to an average score of 500 with a standard deviation of 100. To make a score in one year comparable to a score in another year, all future verbal SAT scores will be linked to this reference curve, via a process called "equating". For example, a student obtaining a score of 600 in one year would be considered equivalent in ability to a student obtaining a score of 600 in any other year. The same reference curve will be used until March, 1995. One effect of equating is the necessity of keeping the SAT content and question types generally the same from one year to the next going forward. Another effect is that average SAT scores are no longer fixed to be 500.

In December, administration of the original College Board examinations is suspended, and the exams are not used again. At this point, the SAT is the standard admissions test for almost all of the private colleges and universities in the northeastern United States.

From this time forward, the SAT is entirely machine scored, using a technique that measures electrical conductivity in the marks made by pencils.

1942

Math returns to the SAT in April, in the form of multiple-choice questions with five-choice answers. To make a score in one year comparable to a score in another year, all SAT math scores on future exams will be linked to the curve used on the math section of this year's April exam. The same reference curve will be used until March, 1995.

1946

The SAT verbal section is changed to consist of antonyms, analogies, sentence completion, and reading comprehension, with somewhat less emphasis on "puzzle-like" reasoning questions and more emphasis on reading skills. This basic format will remain essentially the same for almost the next 60 years. The reading comprehension portions of the test are specifically considered to be "probably non-coachable".

In Brooklyn, New York, Stanley Kaplan begins teaching SAT prep classes. Each class consists of 4 hours of instruction per week for 16 weeks, at a cost of $128 per student. (About $1450 in 2012 dollars.)

1947

The Cooperative Educational Testing Service (ETS) is founded to consolidate development and administration of a variety of tests, including the Graduate Record Examination (GRE), the Medical College Admission Test (MCAT), and the SAT. The ETS assumes the testing activities of the College Board and other related organizations, but the College Board retains ownership and control of the SAT. To this day, the ETS develops and administers the SAT and AP tests as a contractor for the College Board.

1952

Antonym questions on the SAT are changed to multiple-choice form with five possible answers. About half of the testing time allotted to the verbal section is devoted to reading questions at this point. The College Board begins to compute annual average SAT scores among all test takers at this time.

1958

The number of verbal questions on the SAT is reduced from more than 110 to 90. This change moves the SAT away from a test that was designed so that few students could finish. Reading comprehension makes up about 40% of the test at this time.

Students are allowed to view their own SAT scores for the first time.

1959

In the summer, the American College Testing (ACT) Program is founded by Ted McCarrel and E. F. Lindquist. Lindquist suggests that there is a need for a new regional or national test for college-bound high school students, for several reasons: 1) the SAT is used primarily by selective colleges in the northeastern U.S., but not by most public institutions as well as by universities in other regions of the country; 2) the new test should be used not just for admissions but placement as well; and, 3) the test should primarily be useful as an indicator of academic preparation, i.e., it should be an achievement test.

In November, the ACT Assessment is administered for the first time to about 75,000 students, and scheduled to be administered four times per year (in February, April, June, and November) starting in 1960. Based on the Iowa Tests of Educational Development (in fact, all the ACT exam questions for the first test had been pre-tested on a previous ITED), the test is comprised of four sections: English, mathematics, social studies, and natural sciences. Each section is 45 minutes long, for a total test time of 3 hours. Scores are reported on a scale of 0 to 36 for the test as a whole and for each sub-section. The test administration is primarily limited to the midwestern U.S at this time; the student's test fee is $3. The first ACT test and all successive administrations are scored by computers using optical mark recognition, at rates of thousands of test sheets per hour.

A new SAT math question type, "data sufficiency", is added. Each question is accompanied by two statements, and has five possible answers, each of which are statements such as: "BOTH statements (1) and (2) TOGETHER are sufficient to answer the question asked, but NEITHER statement ALONE is sufficient."

The first Preliminary Scholastic Aptitude Test (PSAT) is administered. The test is a shortened form of the SAT and is designed to help students become familiar with the question types and format of the full exam.

1961

The number of students taking the SAT this year is more than 800,000, roughly ten times the number taking the test in 1951. The number of students taking the ACT this year is about 300,000. (For comparison, the total number of U.S. high school graduates in 1960 is roughly 1.6 million.)

1965

The College Board publishes "Effects of Coaching on Scholastic Aptitude Test Scores", also known informally as the "little green book". The book states that coaching for the SAT produces insignificant score increases. (The average increase attributable to coaching is said to be fewer than ten points per section.)

1968

The number of students taking the ACT in the 1967-68 school year reaches about 950,000, more than seven times the number taking the test in 1959-60, the first ACT testing year.

1970

Starting this year, reported SAT scores are rounded to the nearest number divisible by ten. Previously, it was possible for students to receive scaled scores such as 501 or 789, for example.

1971

The National Merit Scholarship Corporation begins co-sponsoring the PSAT, which is now also called the National Merit Scholarship Qualification Test (NMSQT). Scores on the PSAT will be used to determine which students will receive recognition of scholarship and/or scholarship money.

1974

The number of reading comprehension questions is reduced to about 30% of the verbal portion of the SAT, in favor of more antonym and analogy questions.

In the math portion of the SAT, data sufficiency questions are replaced with "quantitative comparison" questions, which have four possible answers. The quantitative comparison questions ask the student to determine whether two quantities are equal, different (and which is larger), or indeterminate.

The total time of the SAT verbal and math portions is reduced from 3 hours to 2 1/2 hours in order to accommodate the half-hour Test of Standard Written English (TSWE) that is newly added to the SAT. The TSWE is scored on a separate scale (20-60) and consists of multiple-choice questions designed to evaluate grammar and writing skills. The results of the TSWE are expected to be used by colleges for the appropriate placement of the test taker in freshman English class.

1975

The U.S. Federal Trade Commission (FTC) begins an investigation into the Kaplan Educational Center, a test preparation company. The FTC is investigating whether Kaplan is making false claims in its advertising. The Kaplan advertisements running in Boston say that, on average, Kaplan students raised their (combined verbal and math) SAT test scores by 100 points.

1977

The number of SAT tests administered per year in the U.S. increases to six.

1978

The College Board begins including an entire sample SAT in its handbook (called "Taking The SAT") given to students. This particular sample test is the first complete SAT to be made public. The previous handbook (called "About The SAT") included sample questions, but not an entire test.

At this time, the test fee for the SAT is $4.75 (about $16 in 2012 dollars).

1979

The FTC releases the final report of its investigation of the ETS, Kaplan, and other test preparation companies. The report suggests that coaching can improve SAT scores on average by 50 points (combined math and verbal). In a re-analysis of the FTC data, the ETS suggests that the result could be due in whole or part to the increased motivation and desire of students who choose to be coached, compared to those who do not.

In July, New York State passes the "Admissions Testing Law", often called the "Truth-In-Testing Law". The law, to take effect in January, 1980, requires that students be allowed to see actual copies of their tests and answer sheets.

In December, the ETS announces that copies of some previously released SATs will be released to students nationally, on an ongoing basis.

1980

The College Board begins selling previously administered SAT tests directly to students, with the release of the booklet "4 SATs", to be followed by "5 SATs" in 1981 and "6 SATs" in 1982. In 1983, the College Board will begin regularly publishing tests in books, available nationally in book stores, called "5 SATs" and "10 SATs".

To comply with the New York truth-in-testing law, the College Board reduces the total number of SAT administrations in that state from eight to four, and increases the SAT fee from $8.25 to $10 for New York students. Each student can receive a copy of the test with his or her answers and the correct answers, for an additional $4.65.

1981

The College Board raises the math scores of nearly a quarter million students who took the PSAT administered in October, 1980. This PSAT is the first to be released to the test takers; a student notices that the ETS answer to one of the math questions (dealing with pyramids) is incorrect.

In March, the College Board decides to provide all students nationally with copies of their SAT exams and answers, for at least some test administrations.

1986

In the 1985-1986 school year, nine students out of about one million test takers (roughly one in 110,000 test-takers) receive a perfect score of 1600 on the SAT.

1989

In October, a new version of the ACT (called the "Enhanced ACT") is administered, replacing the previous version of the test. Two major changes are made: the "Natural Science" sub-test of the ACT is replaced with a "Science Reasoning" sub-test, and the "Social Studies" sub-test is replaced with a "Reading" sub-test. The new reading sub-test is designed to be a better assessment of "pure" reading ability and comprehension, whereas the social studies sub-test contained items testing, among other things, specific knowledge of U.S. history. The new science sub-test de-emphasizes specific scientific knowledge while primarily assessing analytical and problem-solving skills using reading material, charts, graphs, and tables drawn from scientific literature.

With the new ACT test, the scaled scores are also "recentered". Although new scores will still be reported on the same 1-36 scale, the recentering means that the scores for the previous test will not be directly comparable to scores for the new test. The change increases the average composite score from 18.6 for 1989 seniors (old scale) to 20.6 for 1990 seniors (new scale).

At this time, roughly one million students take the ACT each year, whereas about 1.2 million students take the SAT. (For comparison, the total number of U.S. high school graduates in 1990 is roughly 2.5 million. About 40% of all high-school graduates in the U.S. take the SAT in this year.)

1990

In this year, ten students out of 1.2 million test takers (roughly one in 120,000 students) get perfect scores of 1600 on the SAT.

1993

The SAT is renamed from "Scholastic Aptitude Test" to "SAT I: Reasoning Test", and the Achievement Tests are renamed "SAT II: Subject Tests". The first renamed tests will be administered in March, 1994. Collectively, according to the College Board, these tests are to be known as "Scholastic Assessment Tests" (plural), and the acronym "SAT" is no longer considered to stand for anything. However, for at least the next three years, the Reasoning Test is commonly (but incorrectly) called the "Scholastic Assessment Test" (singular).

The president of the College Board says that the renaming is designed "to correct the impression among some people that the SAT measures something that is innate and impervious to change regardless of effort or instruction."

1994

Significant changes are made to the SAT starting with the March test. Antonyms are removed from the verbal section to make rote memorization of vocabulary less useful. The percent of content devoted to passage-based reading material is increased from about 30% to about 50%, and the reading comprehension sub-sections are renamed "Critical Reading". The reading passages are chosen to be more like typical college-level reading material, compared to previous SAT reading passages. The changes for increased emphasis on reading as well as changes to the math section are made in response to a 1990 report issued by a commission established by the College Board. The commission recommended that the SAT should: "do more than predict college grades", "reinforce the growth of sound high school curricula", and "approximate more closely the skills used in college and high school work".

The TSWE is dropped from the SAT at this time; this test reappears as part of the SAT II Writing Test, which also includes a short essay. The time allocated to the math and verbal portions increases by 15 minutes each, keeping the SAT three hours in length.

Three major changes are made to the math section of the SAT: the tested math content is expanded, free-response questions are added, and students are now allowed to use calculators. These changes are made in response to the suggestions of the NCTM, which in an influential 1989 report had emphasized the use of the "real-world" problems, probability and statistics, and calculators in the K-12 math curriculum. At this time, the tested math material is expanded to include: questions with more than one correct answer (via the free-response section); coordinate geometry; slopes of lines; probability; basic statistics such as mean, median, and mode; and, counting problems.

1995

Starting with the April SAT, scaled scores are "recentered". By the early 1990s, the average SAT verbal scores were about 425 and the average SAT math scores were about 475. The table below shows average SAT scores for seniors graduating in the listed class year:

	Average SAT Score (Seniors)
Class Year	Math	Verbal
1952	494	476
1962	495	474
1972	484	453
1982	467	426
1992	476	423
1996	508	505

(For all of the yearly average SAT scores from 1952 to the present, including SAT scores on the original (pre-1995) scale, see this PDF file for the data and this PDF file for a plot.) The recentering is done in order to return the average scores of the verbal and math sections closer to each other and closer to the midpoint of the scale (500), as seen in the last line of the table above. Although new scores will still be reported on the same 200-800 scale, the recentering means that the old test scores (prior to April 1995) will not be directly comparable to later scores. For example, a May 1995 score of 600 in math will not reflect the same ability level as a May 1994 score of 600 in math.

The primary problem with the pre-1995 scale is that test scores are still linked to the 1941 and 1942 reference groups of students, and the test-taking population changed significantly in the decades after World War II. Mathematically, this meant that it was not unusual, especially for verbal section scales, to have a perfect raw score correspond to a scaled score of less than 800. (The scoring policy, however, was to award an 800 for a perfect raw score.) A group of about one million seniors in the class of 1990 is chosen to be the reference group for the new scale.

Another effect of the recentering of SAT scores is a significant increase in the number of students achieving a perfect score of 1600. Previous to the new scaling, a single mistake or question left blank would result in a score of less than 1600. Starting with the April, 1995, SAT test, students can miss as many as four questions and still get a perfect 1600. In 1994, 25 students got perfect scores out of about 1.25 million (about 1 in 50,000 students). The first recentered SAT in April has 137 perfect scores out of about 200,000 test takers (about 1 in 1,400 students).

1996

In September, the ACT (both the test and the company) is renamed so that "ACT" is no longer an acronym: the letters "ACT" no longer stand for anything.

Starting in October, calculators are allowed for use by students on the math section of the ACT test.

1997

The College Board, in an attempt to clear up confusion about the naming of the SAT, says that the SAT by itself is not properly called the "Scholastic Assessment Test". Instead, the term "SAT" is not to be considered an acronym: the letters "SAT" no longer stand for anything.

Online registration for the ACT test via the Internet is made available.

2001

Concerned that the SAT is not nearly as good a predictor of college success as either high-school grades or the SAT II Subject Tests, the president of the University of California suggests dropping the SAT I as a consideration in UC admissions. Criticisms of the SAT at this time also include the apparent disconnection between what high-school students are learning in their course work and "esoteric" items on the SAT such as verbal analogies and quantitative comparison questions.

2002

"Score Choice" for the SAT II Subject Tests is dropped. Previously, a student could decide whether a Subject Test score would be sent to a college or university. After the change, all scores of any tests taken are sent, matching the policy of the SAT I.

2003

The last version of the "10 SATs" books (the third edition of "10 Real SATs") is published by the College Board. A similar book (with ten or even five previously administered SATs) has not been published since. (As of 2012.)

2004

The SAT is again renamed, dropping the roman numerals, so that the official names are "SAT Reasoning Test" and the "SAT Subject Tests".

2005

Beginning with the March SAT, the content of the test is changed, at least partly in response to the UC criticisms. The "Verbal Reasoning" section of the SAT is renamed "Critical Reading", and the verbal analogy questions are dropped. Newly added is a writing skills section, with essay, based on the now discontinued SAT Subject Writing Test. Three SAT scores, for Critical Reading, Math, and Writing, each on a scale of 200-800, are reported, making the perfect score 2400 instead of 1600.

In SAT Math, quantitative comparison questions are dropped. Several new topics are added: exponential growth; absolute value; functional notation; equations of lines; data interpretation, including scatterplots; rational and radical equations; and, manipulation of fractional and negative exponents. (The rational and radical equations as well as the fractional and negative exponents are added to reflect content from typical third-year high-school algebra courses.) Greater emphasis is placed on linear functions, and properties of tangent lines.

The total time of the SAT (including a 25-minute equating section) increases to 3 hours and 45 minutes.

About 300,000 students take the first "new" SAT in March, with 107 of them (roughly 1 in 2,800 students) receiving a perfect score of 2400.

The ACT test adds a writing section, beginning with the February administration; the section is optional for test takers.

2006

Out of 1.38 million seniors, 238 (roughly 1 in 5,000 students) receive a perfect score of 2400. In 2004, roughly the same number of seniors took the SAT, and 939 (roughly 1 in 1,500 students) received a perfect score of 1600.

2007

The ACT becomes a valid admissions test at every four-year college or university in the U.S. when Harvey Mudd College accepts ACT scores for fall admissions.

2009

"Score Choice" for the SAT is returned: Students are allowed to report any or all of the SAT or SAT Subject Tests that they take, depending on the score policy of the recipient colleges. (However, the honor system is used: no verification is made by the College Board that a student reports all scores to a college that has an "all scores" policy.) Previously, all SAT and SAT Subject scores would be reported.

2010

For the first time since the ACT test has been administered, the number of high school seniors taking the ACT (1.57 million) is greater than those taking the SAT (1.55 million).

Notes

Answer to the 1926 verbal questions: carrot, potato, and beet; sincere and dissembling.

Answer to the 1926 math questions: sequence - 15 and 90; boat - 3 hours, 20 minutes.

This history is a little too SAT oriented. I need a little more ACT-bell!

References

"The Scholastic Aptitude Test: Its Development and Introduction, 1900-1948", David H. Rubin, Ph.D. Dissertation, University of Oregon, 1988.

"Beyond Prediction", Commission on New Possibilities for the Admissions Testing Program, College Board, 1990.

"A Historical Perspective on the Content of the SAT", College Board, 2003.

"The First 75 years: The College Board 1900-1975: A Historical Perspective", College Board, 1975.

"A Faithful Mirror", College Board, 2001.

"The Recentering of SAT Scales and Its Effect on Score Distributions and Score Interpretations", College Board, 2002.

"The Math Section on the New SAT", College Board, 2004.

"ACT: The First Fifty Years, 1959-2009", ACT Inc., 2009.

"Reflections on a Century of College Admissions Tests", Atkinson and Geiser, 2009.

"Digest of Education Statistics: 2009", U.S. Dept. Of Education / Institute of Education Sciences, 2009.

"On Further Examination: Report of the Advisory Panel on the Scholastic Aptitude Test Score Decline", College Board, 1977.

"Secrets of the SAT", PBS Frontline, 1999.

"When Close Is Perfect: Even 4 Errors Can't Prevent Top Score on New S.A.T.", James Barron, The New York Times, July 26, 1995.

"Picture-Perfect Score for Teen: A 1600 on SAT", Greg Hernandez, Los Angeles Times, January 31, 1991.

"No Mistake, These Five Kids Have All the Answers: They Scored Perfect 1600s on Their SAT Exams", Michael Small, People Magazine, June 8, 1987.

"Effects Of Coaching On Standardized Admission Examinations (Revised Analysis)", Federal Trade Commission, 1979.

"The Effectiveness of Coaching for the SAT: Review and Reanalysis of Research from the Fifties to the FTC", Samuel Messick, ETS, 1980.

"Test Pilot: How I Broke Testing Barriers for Millions of Students and Caused a Sonic Boom in the Business of Education", Stanley H. Kaplan, Simon and Schuster, 2001.

"Youth Outwits Merit Exam, Raising 240,000 Scores", Edward B. Fiske, The New York Times, March 17, 1981.

"'Truth In Testing' To Be Nationwide", Edward B. Fiske, The New York Times, March 27, 1981.

"Q. & A. on 'Truth in Testing'", Dan Hulbert, The New York Times, January 6, 1980.