The ACT is better at measuring some aspects of student success than others.

Approximately 300,000 students take the American College Test each year. Many colleges and universities require prospective students to take the ACT or the SAT to measure their potential to do well in college. But is the ACT a good measure of future college success? Some experts say the ACT is an accurate predictor of a student's good work ethic in college. However, other experts say the test is biased.

Objectives of the Designers

The American College Testing Program developed the ACT as a response to the College Board's Scholastic Assessment Test in 1959. According to the ACT website, the test was created because colleges wanted more than one test to assess a student's potential for college success. The ACT has four sections: English, reading, math, science, and an optional writing component. The English component includes grammar and mechanics, and the reading portion includes reading comprehension and inferences. Math test questions include questions in algebra, geometry and trigonometry. Science test questions ask examinees to interpret, and analyze test questions and data. According to the ACT website, test designers hoped that the ACT would measure a student's general educational development and their ability to complete work on the college level.

What the ACT Measures Well

According to an article in the "Washington Post," a study of the ACT found that the ACT's mathematics and English subtests have been found to be fairly good predictors of college success in classes in these subjects. In fact, studies have shown that the math and English tests by themselves do a better job at predicting college success than the ACT as a whole. Some researchers have said that these subtests do a better job predicting college success because they test a wide range of skills. The American College Testing Service says their ACT composite score is a good predictor of college success, and researchers agree -- if a student's scores on the four subtests are relatively close together.

What the ACT Doesn't Measure Well

Some research has shown that the ACT's subtests in reading and science are not an accurate predictor of college success. According to the ACT website, the reading test has the examinee look for main ideas and details, compare and sequence stories and events, identify and understand cause and effect relationships, and find the meanings of words and phrases in context. The student may be tested on passages in social studies, science, fiction or nonfiction. The reading level of the ACT is estimated to be at about a 10th grade level or above. The science test assumes that the student has finished courses in physical science and biology and can answer questions from multiple sources of data such as tables and graphs.

Bias on the ACT

Critics of standardized testing, such as the Fair Test organization, charge that tests like the ACT are biased against minorities and women. Women score an average of two to three points lower overall on the ACT than men, according to Fair Test. Women also score significantly lower than men on the math and science subtests. Part of the issue, according to a study done by the Educational Testing Service, lies in the use of the multiple choice question format. When short answer, fill-in-the-blank and essay formats are used along with multiple choice format, the gender bias on the test disappears entirely, according to the study. In addition, according to Fair Test, the ACT uses language such as idioms that may not be understood by all test takers because it comes from only one culture and is not representative of all cultures, especially for minorities and second language learners. This bias, say critics, makes the ACT and other tests an inaccurate measure of college aptitude. For these students, high school grades and AP test scores may be a better predictor of college success.