reliability and validity of standardized tests

Tests need to be fair to be valid, free from bias, and ideally fit all test-takers, independent of their gender, race or socio-economic background. The report further argues that colleges would get more successful students if they examined the predictive value more closely.In recent years, there has been an increasing focus on the need to consider not only the validity and reliability of tests but also their fairness. In a study completed in 2011 and published in the May 2013 issue of the American Economic Journal, researchers from both Stanford University and the University of Chicago question the predictive validity of the science and reading portions of the ACT. The ACT Plus Writing has the four area multiple-choice tests plus a Writing Test. The ACT, first given in 1959, is a college readiness assessment that measures a student’s readiness for college in the areas of English, mathematics, reading and science. The subtests are better predictors on whether or not a student will be successful in specific classes in college. Richard Atkinson, former chairman of the Board on Testing and Assessment for the National Council of Research, noted that the SAT is an exam at war with itself because the core test is not as valid as the optional subtests. According to the College Board, publisher of the exams, the SAT “is consistently shown to be a valid predictor of college success for students from all backgrounds,” but in a 2012 speech, Dr. The basic skills of reading, mathematics and writing form the core of the SAT however, there are 20 subtests in the areas of foreign language, mathematics, history and science that can also be examined. Since 1926, the SAT has been used as a predictor of college success in those subjects tested by the exam. Consequently, how well a student does on the SAT or ACT should predict how well he will do in college. This refers to the standardized test being able to “predict” an outcome based on assessment responses. The criterion-related validity in testing can be further broken down and discussed as predictive validity. To examine the validity of the ACT and SAT, researchers look specifically at criterion-related validity, which refers to the fact that a student has to show mastery of certain criteria or data that has been learned. Four methods are used to gauge the validity of an exam: content validity, criterion-related validity, construct validity and, most recently, consequential validity. Validity in educational testing refers to whether or not a test “examines” or measures what it says it does.

Since the scores of these college admission tests have become a funding opportunity for the student, the question remains: What exactly is being examined? That is basic validity.

Reliability and validity of standardized tests