Standardized Tests Are Not a Good Predictor of College Success - Research Paper Example

Standardized Tests Are Not a Good Predictor of College Success Introduction The history of standardized tests takes one to ancient China where imperial examinations covered the six arts which included music archery and horsemanship, arithmetic, writing and knowledge of rituals and ceremonies (Hopkins 36-47; Ransdell 357-364). With time, standardized tests spread to other parts of the world and presently, they have been adopted in many countries. A standardized test refers to a test that is administered and scored in a consistent or standard manner (Hoffman and Katie 455-474). They are designed in such a way that there is consistence in the questions, conditions for administering, scoring procedures and interpretations (Hopkins 36-47). Standardized testing in the United States is a phenomenon of the 20th century with origins in the First World War, the Army Alpha and Beta tests developed by Robert Yerkes and colleagues (Hoffman and Katie 455-474). It is important to note that standardized testing is not only applied in schools but also in workplaces’ to evaluate the workers’ ability and in many other areas. Examples of standardized tests in the United States include ACT (American College Testing), SAT (Scholastic Aptitude Test), ISEE (Independent School Entrance Examination), HSPT (High School Placement Test), IQ tests such as the Stanford-Binet Intelligence Scales and Wechsler Adult Intelligence Scale (Hoffman and Katie 455-474; Ransdell 357-364; Hopkins 36-47). Standardized tests have the main objective of evaluating students’ or workers’ understanding, comprehension, knowledge and capability in a particular area. This essay focuses on standardized tests that are used as determinants in college entry which include ACT (American College Testing) and SAT (Scholastic Aptitude Test). Much as standardized tests are used by colleges to admit students, they are not a good predictor of college success. This paper explains why standardized tests cannot be used to forecast the student’s performance and eventual success in college. Standardized Tests and College Success For a better part of their history, standardized tests have received criticism as faulty instruments for predicting college success. Among other factors, standardized tests disfavor high-order learning, favor rich students at the expense of poor ones, put schools at the risk of producing lower test scores with negative political consequences and poor performance due to anxiety and lack of time management (Ransdell 357-364; Hoffman and Katie 455-474). Standardized tests cannot measure initiative, creativity, imagination, curiosity, judgment, ethnic reflection and a host of other dispositions and attributes (Hopkins 36-47). This is for the reason that they can only measure isolated skills, specified facts, specific functions, content knowledge and other least interesting and less significant learning aspects (Hoffman and Katie 455-474). For these and other reasons, it would be fallacious to predict that a student will succeed in college on the basis of their performance in the standardized tests. There is a temptation to use standardized tests to define the school curriculum and focus on instructions and resources towards the same. It is important to note that the main objective of education, the world over, is to create resourceful members of the society who can change their status from bad to good and good to better (Ransdell 357-364). For this reason, education systems should be geared towards instilling skills necessary to attain this objective which calls for a curriculum that is determined by the same noble objective. It is unfortunate that with standardized tests, this is not case. It is the standardized tests that determine what is taught in class and what is not since they change the goal of schooling to passing the tests and not attainment of skills (Hopkins 36-47). What happens is that students reproduce what they are taught in class through cramming and they score well and join college. The question is not about joining college but whether or not these students who are used to cramming will succeed in college. It is possible that a school can efficiently and effectively use standardized tests without letting its limits control the school curriculum and classroom instruction. However, this can, to a huge extent, be detrimental to the school by exposing it to the risk of producing low test scores in cases where the school may not teach in alignment with standardized testing. This can translate into grave political consequences for the school as stipulated in laws such as the federal No Child Left Behind in the United States of America (Ransdell 357-364). This means that test scores are used as means of accountability where schools that are labeled “in need of improvement” can be punished. In such instances where standardized testing is used as the sole method of accountability, parents and the community find it challenging to know the students’ performance in areas that are not tested (Hopkins 36-47). Well and good that students may pass in SAT and ACT and join college. But this does not necessarily mean that they will perform well in college. This could best be accounted for by the fact that even the students themselves may not know their strengths and weaknesses and they may end up specializing in areas that they are weak which may eventually lead to college failure. This, in itself, is a clear indicator that standardized testing is not a good predictor of college success. Each and every college weighs standardized test scores differently. More than 400 colleges do not require submission of ACT and SAT tests scores on admission (Ransdell 357-364). Colleges do not emphasize on the use of standardized test scores for the reason that the tests are designed to predict first-year college grades rather than predicting grades beyond freshman year, graduating rates or pursuit of graduate degree. Correlation coefficient is generally used to examine the relationship between test scores and first-year college grades. Research indicates that predictive ability of SAT on freshman’s college grades is just 0.22 which explains only 22 per cent of the variation in freshman college grades (Ransdell 357-364; Hopkins 36-47). A correlation of high school grades against freshman college grades stands at 0.54 which translates to 54 per cent and a variance of 32 per cent in the performance of first year in college. This shows that high school grades rather than test scores are best placed to predict college success. According to Baron and Norman, there is a weak predicative ability of SAT and ACT on college success (37-69). Baron and Norman (37-69) looked at the relationship between student’s performance in SAT and cumulative college grades as compared to the relation between class rank cumulative college grades. It was found that SAT I was the weakest predictor with on only 4 percent of the variation in college grades while SAT II followed with 6.8 percent of differences in academic performance. Class rank was found to be the most useful tool that predicted 9.3 percent of the changes in cumulative college grades. As a matter of fact, combining class rank and SAT I summed up to 11.3 percent which leaves unexplained 90 percent of the variation in college grades (Baron and Norman 37-69). Vars and Bowen conducted a study to determine the test score gap between black and white students (57-67). Among other things, the study entailed validity research to examine the relationship between SAT scores and college grades. They conducted a study of 10,000 students at 11 selected public and private higher institutions of learning. Holding race, gender and field of study constant, it was found that there was a 100-point increase in SAT combined scores which led to only one tenth of a grade point gain for college GPA (Vars and Bowen 57-67). This shows that there is a tenuous correlation between standardized test scores and college success. President Richard Atkinson of the University of California conducted a three year validity study to analyze the power of SAT I , SAT II and high school grades as predictors of good performance at California’s eight public universities (Ransdell 357-364). The validity study by the University of California tracked 80,000 students from the year 1996 to 1999 and it showed a poor weakness in terms of predictive power of SAT I which accounted for only 12.8 percent of the variation in GPA (Bowen and Derek 46-59). SAT II accounted for 15.3 percent of the variation in GPA. The study also found out that SAT I scores were more susceptible to the influence of the socio-economic status of the students which further compliments the statements for criticism of standardized testing. Crouse and Dale look at the situation beyond college grade and indicate that SAT is a poor utility in predicting long-term success by students in college (45-57). According to Crouse and Dale, SAT is a poor predictor of college success but other unrecognized predictors are (45-57). They, therefore propose dropping SAT in forecasting long-term college success and relying on readily available measures such as students’ high school coursework and grades and use of achievement tests that measure the mastery of high school courses. In particular, their study shows that the use of high school record alone in predicting the student who will complete a bachelor’s degree resulted to 73.4 percent of all cases (Crouse and Dale 45-57). In fact, if the high school GPA forecast is used together with SAT I, the number comes down to 72.2 percent of the same cases. This is a clear indicator of the weak correlation that lies between standardized test scores and college success. If performance of males is compared to that of females, then it becomes clear how poor standardized tests are in predicting college success (Bowen and Derek 46-59). There are long outstanding gaps between males and females, holding other factors constant, on the overall performance of SAT. On average, females score 35 to 45 points lower than males in SAT. A closer look at high school and college grades are indicative of a different picture. This is for the reason that females perform better on the basis of high school and college grades as compared to their male counterparts and yet the latter are better performers in SAT (Hopkins 36-47). This shows that the test consistently over-predicts the performance of males and under-predicts that of females. This is a clear indication that good performance in standardized tests such as SAT does not in any way guarantee college success. On the basis of the language factor, standardized tests are biased (Bowen and Derek 46-59). There exist students who do not have English as their first language. Frequently, standardized tests under-predicts these students and their future performance in college. At times, these students are left out during admissions into colleges in cases where they perform poorly due to low understanding of the language. A study by the University of Miami made a comparison between Hispanic and non-Hispanic white students (Bowen and Derek 46-59; Ransdell 357-364). The performance of the two groups was equivalent at the college level. However, the average combined SAT scores of the Hispanic students were 91 points lower than their non-Hispanic white counterparts. Due to language difficulty amongst Hispanic students, their performance in SAT was relatively lower than that of non-Hispanic white students who are favored by the language factor in SAT. It is evident that, between the two groups, there existed a gap in the performance in SAT but none existed in the performance in college which compliments the idea that standardized tests are not a good indicator of college success. SAT and ACT do not put into consideration the equity question of education as they have questionable predictive power for poor students. This is for the reason that the ability of SAT and ACT to predict freshman grades, undergraduate class rank, college graduation rates and attainment of a graduate degree is weaker for poor students than those with rich backgrounds. This can be explained by socio-economic status and the level of education of parents for both groups which is higher for the latter as compared to the former (Bowen and Derek 46-59). Most of the time it has been alleged that inequality is vested in preparation of these standardized tests which ultimately affect their outcomes. Rich parents can afford to pay for out-of-school coaching and buy sample test papers for their children which may give the rich students an advantage over the children from less affluent families. In fact a study by the University of California found out that SAT and ACT scores proved to be susceptible to the influence of socio-economic status of applicants (Hopkins 36-47). Poor students might perform poorly as compared to their rich counterparts in SAT and ACT yet they have the capacity that was not adequately tapped due to lack resources such as finances to prepare well for the tests. When they join college, as the study found out, the poor students perform better than those from affluent backgrounds. Such discrepancies put into question the relevance of standardized testing to predict college success (Bowen and Derek 46-59). Conclusion In conclusion, the increasing number of colleges that admit substantial numbers of applicants without putting into consideration the standardized test scores is indicative of the fact that college success cannot be well predicted on the basis of standardized tests. Other predictors have proved their effectiveness in forecasting college success and they include, but are not limited to, class rank, high school grades and the level of seriousness in classroom. As a matter of fact, the weak predictive power of SAT, its susceptibility to teaching, test score misuse and the negative effects of standardized test score on equity on education are good and enough pointers that standardized tests should not be used as a criterion to admit students into colleges. This is a good point to demystify the fallacy that the student’s performance in standardized tests is an indicator of college success. To that effect, standardized tests are not a good predictor of college success and they should therefore, not be used at all. Works cited Baron, J. & Norman, M. F. Education and Psychology Measurement. University of Pennsylvania, 1992: 37-69. Bowen, G. & Derek, B. The Shape of the River. Princeton University Press, 2000: 46-59. Crouse, J. & Dale, T. The Case Against The SAT. University of Chicago Press, 1988: 45-57. Hoffman, J. & Katie, L. “Predicting College Success with High School Grades and Test Scores: Limitations for Minority Students.” The Review of Higher Education. 28.4 (2005): 455-474. Hopkins, K. Teaching How to Learn in a What-to-Learn Culture. John Wiley and Sons, 2010: 36-47. Ransdell, S. “Predicting College Success: The Importance of Ability and Non-Cognitive Variables.” International Journal of Educational Research. 35.4 (2001): 357-364. Vars, F. & Bowen, W. The Black-White Test Score Gap. 1998: 57-67. Read More