Performance of the USA in the Scholastic Assessment Test - Research Paper Example

PERFORMANCE OF THE UNITED S OF AMERICA IN THE SCHOLASTIC ASSESSMENT TEST INTRODUCTION Background of the Study Aptitude tests are examinations that elicit information from the examinees, usually students who want to pursue a college degree. The results of these tests give predictive information on the performance of the students in their planned field of endeavor (McDonald et al, 2001). One of the most widely known and intensively examined aptitude test for higher education is the SAT. This was first used in the United States in 1926 and was called Scholastic Aptitude Test. However, in 1994, a major revision was made on test which was directed to suit the changing needs; it was also renamed and called Scholastic Assessment Test (The Sutton Trust, 2000). The SAT gives two scores, the mathematical and verbal scores. Each of these scores is reported as a scaled score which has a mean of 500 and a standard deviation (a measure of the spread of scores) of 110. It is also known that the maximum score for each part of the SAT is 800 and the minimum 200. By the use the above-mentioned scaled scores, it is possible to obtain an indication of how the verbal or math scores for an individual student compare with the scores of others, as this transforms all scores into an approximately normal distribution. Since it approximates a normal distribution, by empirical rule, about 68% of the students who will be taking the test will have a math and verbal score that lies within one standard deviation either side of the mean (that is, between 390 and 610) and 95% will have a score within two standard deviations of the mean (that is, between 280 and 720). Only 2.5% of test takers would be expected to score less than 280, with a further 2.5% scoring above 720 (Allalouf and Ben-Shakhar, 2001). Since 1926, the SAT was administered annually and only in 1994 that a major revision was made, including its meaning. The major changes to the SAT were documented by Steven Graff (1993), project director of the working group responsible for the redevelopment of the SAT. Prior to 1994 the verbal section of the SAT consisted of: • Analogies measuring reasoning skills and knowledge of vocabulary (20 questions); • Sentence completions primarily measuring logical relationships among parts of a sentence (15 questions); • Antonyms measuring knowledge of vocabulary (25 questions); • five or six reading passages of between 200 and 450 words each, with questions assessing inference, application or evaluation of logic or style, and questions relating to the main idea of the passage. Passages were drawn from a range of subject areas, and included at least one passage that was considered ‘minority-relevant’ (25 questions). These questions were organized in two 30-minute sections. A 30-minute Test of Standard Written English (TSWE) was also included as part of the verbal test. The major changes to the revised SAT involved dropping the antonym section and replacing the TSWE with a separate test which also included English composition - the SAT II: Writing Test. The emphasis on the assessment of reading was also increased in the redevelopment. This was done as reading, particularly the ability to read critically, was seen as being of central importance to any course of study at college. The revised, current verbal SAT consists of: • Analogies measuring reasoning skills and knowledge of vocabulary (19 questions); • Sentence completions measuring logical relationships and vocabulary in the context of the sentence (19 questions); • Four reading passages of between 400 and 800 words each, with questions assessing extended reasoning skills, literal comprehension and vocabulary in context. Two of the passages are paired and complement each other in some way (e.g. represent differing sides of an argument), and so support questions relating to differing styles of writing or points of view, and one passage is described as being ‘minority-relevant’ (40 questions). The above-mentioned background information regarding the Scholastic Assessment Test (SAT) gave the researcher an interest on how the 51 States of America performed in the said test after the major revision in 1994. Thus, this study was conducted. Statement of the Problem This study sought to investigate the performance of the United States of America in the Scholastic Assessment Examination. Specifically, this answers the following: 1. What is the performance of USA in the Scholastic Assessment Test? 2. Is there a significant difference in the performance of the states that require competency test with those that don’t? 3. Is there a significant difference in the performance of the States of when grouped according to political divisions? 4. What are the significant predictors of SAT performance? Assumptions Some of the assumptions used in this study include the assumption of normality of the scores or performance of the states in the Scholastic Assessment Test; however, there are differences in them that are caused by number of factors (Bridgeman et al, 2000). Definitions of Terms: The following terms are defined based on how they were used in this study. SAT Average combined verbal/math scores of Scholastic Assessment Test during 1994-95 school years. EXPEND current expenditure for public elementary and secondary education for the 1994-95 school years, in billions of dollars. TAKERS Percentage of 1994-95 school year graduates who took the SAT ENROLL Total enrollment in public elementary and secondary school in the United States; fall 1995 figure in thousands of students. TEACHERS Total number of public elementary and secondary teachers in the United States; fall 1995 figures in thousands of full time equivalent RATIO Student-to-teacher ratio in the United States, fall 1995. Calculated as ENROLL/TEACHERS SALARY Estimated average annual salary of teachers in thousands of dollars. TEST DESIGNS This study utilized the descriptive method using co-relational design. This method was chosen because the primary purpose of this study was to identify and describe a certain event, that is the performance of the United States of America in the Scholastic Assessment Test (SAT). According to Bailey (2001) descriptive co-relational research design is appropriate to use when the study aims to describe the nature of a phenomena and its relationship with other variables that might have caused its characteristics. For the statistical tools used in this study, the descriptive statistics were used like mean and standard deviation to answer the problem number 1. Likewise, the t-test for two independent samples was used to find answers for problem number 2, on the comparison of the states according to their requirement related to competency test. Also, the one-way analysis of variance (ANOVA) was utilized in comparing the states as they were grouped according to political divisions. Lastly, a multiple regression analysis and correlation were utilized to look for the significant predictors of SAT performance which include student-teacher ratio, number of teachers, teachers’ salary and a lot more. The researcher used Statistical Package for Social Sciences (SPSS) in the analysis of the data gathered. Sources of Data The study made use of the 1994-1995 SAT results. This was chosen because this is the first results published after the major revision of the Scholastic Assessment Test in 1994. The data was generated from Education Resources Information Center of the Department of Education (USA) official website. RESULTS This portion shows the results of the study to answer the problems presented and to test the hypotheses of this study. Figure 1 Distribution of the States according to Political Regions N=51 Figure 1 shows the distribution of the States according to Political divisions or regions. It is shown in the table that 33% of the states belong to South region, 25% are in the West region, 24% to the Midwest region, and 18% belong to the Northeast region. Figure 2 Distribution of the States According to Their policy on Competency Test Requirement for Graduation N=51 Figure 2 shows the distribution of the states according to their policy on competency test requirement. It is shown that 34 out of 51 or 67% of all the states do not require competency test for graduation while 17 out of 51 or 33% do require competency test for graduation. Figure 3 Histogram of the SAT Performance Figure 3 shows the over-all performance of the United States of America in the Scholastic Assessment Test, CY 1994-1995. It is shown that the mean SAT score is 963.78 with standard deviation of 75.62. Also, it was found that the coefficient of skewness is 0.25. Generally, based on the assumption that the mean is 500 (Allalouf and Ben-Shakhar, 2001), it can be said that the performance of the USA in the SAT was very high. Likewise, the coefficient of skewness also means that the distribution is normal as it can glean from the histogram. Comparison of the States according to SAT Performance Table 1 SAT Performance according to Competency Test Requirement TYPE Mean Std. Deviation t sig require 944.65 74.72 -1.286 0.204 do not require 973.36 75.33 The table (Table 1) shows the performance in the SAT of the states which require competency test for graduation as compared to those that do not. It is shown that the mean score of those which don’t require competency test is 944.65±74.72 while those that do not require scored a mean of 973.36±75.33. By the use of the t-test for independent samples, it can be inferred that the difference is not significant as reflected by the computed t-value of -1.286 with a p-value of 0.204. Thus, there is no significant difference in the performance of the states that require competency test and those that don’t. Table 2 SAT Performance According to Political Divisions/Regions Mean Std. Deviation F Sig Northeast 900.56 15.67 Midwest 1043.75 61.44 11.613 .000 South 940.76 74.04 West 963.85 52.45 Table 2 shows the performance of the different regions of USA in the Scholastic Assessment Test. It is shown that the Midwest region performed the highest with a mean of 1043.75±61.44, followed by the West region (963.85±52.45), South with a mean of 940.76±74.04 and the Northeast (mean=900.56±15.67). The performances of the states were tested for the differences using the Analysis of Variance (ANOVA). It was found that there is significant difference in the performance of the states as grouped according to region as reflected by the F value of 11.613 with a sig value of 0.000. Table 3 Regions Significantly Different In SAT Performance Groups Mean Difference F Sig Midwest vs. Northeast 143.1944 26.06487 .000 Midwest vs. South 102.9853 22.28649 .000 Midwest vs. West 79.9038 23.66276 .016 Table 3 shows the result of the Sheffe’ Test for multiple comparisons. It was found that the performance of the Midwest is significantly different from all the other regions as reflected by the sig. values that are less than 0.05. Therefore, the Midwest region performed significantly greater in the SAT than all other regions in the United States of America. Table 4 Correlation of the SAT Performance with Other Variables Ratio Expend Takers Enroll Teachers Salary Pearson Correlation .092 -.271 -.884** -.229 -.249 -.421** Sig. (2-tailed) .519 .054 .000 .106 .078 .002 N 51 51 51 51 51 51 Table 4 shows the coefficients of correlation of the SAT with all the other variables. The Pearson Product Moment Correlation Coefficient was used. The table shows that the SAT is significantly correlated with the Percent of graduates who took the SAT and the Salary of the Teachers. It can be seen in the table that the relationships are negative. The SAT is highly correlated with the percent of graduates that took the test (r=-0.884) and a low correlation with the Salary of teachers (r=-0.421) (Hurlburt, 2001). These means that the lesser the percent of graduates that took the SAT, the higher their mean performance. Also, it can be interpreted that the lesser the salary of teachers, the higher the performance in the SAT. Table 5 Multiple Regression Analysis Model Unstandardized Coefficients R R squared Sig. B Std. Error (Constant) 1053.184 8.406 .884 .781 .000 Takers -2.512 .190 .000 The table shows the results of the multiple regression analysis. It was found that only the percent of graduates who took the SAT was considered significant predictor of SAT performance. This is reflected by the r-squared or coefficient of determination of 78.1%. This means that 78% of the changes in the SAT rating are attributed from the percent of graduates who took the test. Also, the equation of prediction was found to be: SAT = 1053.184 – 2.512(Takers) References ALLALOUF, A. and BEN-SHAKHAR, G. (2001). The effect of coaching on the predictive validity of scholastic aptitude tests, Journal of Educational Measurement, 35, 1, 31 - 47. Bridgeman, B., Mccamley-Jenkins, L. And Ervin, N. (2000). Predictions of Freshman Grade-Point Average from the Revised and Recentered SAT I: Reasoning Test (College Board Research Report No.2000 - 1/ETS Research Report 00 - 1). New York, NY: College Board. GRAFF, A.S. (1993). The new SAT: the future of transition assessment, Education Libraries, 17, 3, 7 - 13. Hurlburt, Russell (2001) Understanding Behavioral Statistics Brooks/Cole Publishing Company, NY USA THE SUTTON TRUST (2000). Entry to Leading Universities. Executive Summary [online]. Available: http://www.suttontrust.com/text/Report1.doc [29 September, 2000]. Education Resources Information Center (USA) http://www.eric.ed.gov/ERICWebPortal/search/detailmini.jsp?_nfpb=true&_&ERICExtSearch_SearchValue_0=ED395986&ERICExtSearch_SearchType_0=no&accno=ED395986 Appendices Tests of Between-Subjects Effects Dependent Variable:SAT Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 121723.404a 3 40574.468 11.613 .000 Intercept 4.487E7 1 4.487E7 12840.919 .000 Region 121723.404 3 40574.468 11.613 .000 Error 164215.223 47 3493.941 Total 4.766E7 51 Corrected Total 285938.627 50 a. R Squared = .426 (Adjusted R Squared = .389) Multiple Comparisons SAT Scheffe (I) Region (J) Region Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval Lower Bound Upper Bound Northeast Midwest -143.1944* 26.06487 .000 -218.7694 -67.6195 South -40.2092 24.36681 .445 -110.8606 30.4423 West -63.2906 25.63163 .122 -137.6093 11.0282 Midwest Northeast 143.1944* 26.06487 .000 67.6195 218.7694 South 102.9853* 22.28649 .000 38.3657 167.6048 West 79.9038* 23.66276 .016 11.2938 148.5139 South Northeast 40.2092 24.36681 .445 -30.4423 110.8606 Midwest -102.9853* 22.28649 .000 -167.6048 -38.3657 West -23.0814 21.77821 .772 -86.2273 40.0644 West Northeast 63.2906 25.63163 .122 -11.0282 137.6093 Midwest -79.9038* 23.66276 .016 -148.5139 -11.2938 South 23.0814 21.77821 .772 -40.0644 86.2273 Based on observed means. The error term is Mean Square(Error) = 3493.941. *. The mean difference is significant at the 0.05 level. RAW DATA Region STATE SAT EXPEND TAKERS ENROLL TEACHERS RATIO SALLARY TYPE 3 Alabama 1029 3.162 8 735.95 44.1 16.7 31.3 1 4 Alaska 934 1.173 47 125.26 7.4 16.9 49.6 0 4 Arizona 944 3.394 27 766.5 39.6 19.4 32.5 0 3 Arkansas 1005 1.497 6 454.28 29.2 15.6 29.3 1 4 California 902 26.842 45 5447.85 225 24.2 42.5 1 4 Colorado 980 3.338 29 656.28 35.2 18.6 35.4 0 1 Connecticut 908 4.094 81 514.63 36 14.3 50.4 0 3 Wshington D.C 897 0.753 68 108.46 6.5 16.7 40.5 0 3 Delaware 857 0.503 53 79.8 6.4 12.5 43.7 0 3 Florida 889 11.767 48 2172.79 116.8 18.6 33.3 0 3 Georgia 854 6.271 65 1311.13 80.2 16.3 34.3 1 4 Hawaii 889 0.808 57 186.57 10.2 18.3 35.8 1 4 Idaho 979 1.031 15 243.1 12.8 19 30.9 0 2 Illinois 1048 10.419 13 1927.52 113 17.1 41 0 2 Indiana 882 5.67 58 980.2 55.6 17.6 37.8 0 2 Iowa 1099 2.728 5 502.3 32.5 15.5 32.4 0 2 Kansas 1060 2.588 9 464.09 30.9 15 35.5 0 3 Kentucky 999 3.683 11 638.63 37.2 17.2 33 0 3 Louisiana 1021 3.499 9 781.14 47.6 16.4 26.8 1 1 Maine 896 1.282 68 219.23 14.6 15 32.9 0 3 Maryland 909 5.238 64 805.58 46.2 17.4 41.2 1 1 Massachusetts 907 6.302 80 910.02 61.6 14.8 43.8 0 2 Michigan 1033 10.14 11 1643.1 80.8 20.3 49.2 0 2 Minnesota 1085 5.788 9 835.42 48.4 17.3 36.9 0 3 Mississippi 1036 2.015 4 503.6 28.8 17.5 27.7 1 2 Missouri 1045 3.879 9 873.64 57.7 15.1 33.3 0 4 Montana 1009 0.909 21 165.5 10.1 16.4 29.4 0 2 Nebraska 1050 1.498 9 289.73 20 14.5 31.5 1 4 Nevada 917 1.247 30 265.04 13.7 19.3 36.2 1 1 New Hampshire 935 1.195 70 190.45 12.2 15.6 35.8 0 1 New Jersey 898 11.72 70 1197.56 86.9 13.8 47.9 1 4 New Mexico 1015 2.065 11 328.46 19.3 17 29.3 1 1 New York 892 24.6 74 2830 185.9 15.2 48.1 1 3 North Carolina 865 6.114 60 1165.39 71.8 16.2 30.6 1 2 North Dacota 1107 0.552 5 119.09 7.5 15.9 27 0 2 Ohio 975 10.26 23 1838.41 104.2 17.6 37.8 0 3 Oklahoma 1027 2.331 9 616.5 39.3 15.7 28.9 0 4 Oregon 947 3.028 51 527.91 26.7 19.8 39.7 0 1 Pennsylvania 880 12.916 70 1801.97 104.6 17.2 46.9 0 1 Rhode Island 888 1.098 70 148.08 10.2 14.5 42.2 0 3 South Carolina 844 2.97 58 637.52 39.2 16.3 31.6 1 2 South Dakota 1068 0.658 5 144.11 9.3 15.5 26.3 0 3 Tennessee 1040 3.763 12 880.96 50.7 17.4 33.5 1 3 Texas 893 19.33 47 3740.26 240.2 15.6 32 1 4 Utah 1076 1.738 4 473.67 20.1 23.6 30.5 0 1 Vermont 901 0.682 68 105.96 7.7 13.8 36.3 0 3 Virginia 896 5.957 65 1079.85 73.3 14.7 34.7 0 4 Washington 937 5.585 48 951.7 45.3 21 38 0 3 West Virgina 932 1.863 17 306.45 20.9 14.7 32.2 0 2 Wisconsin 1073 5.755 9 869.17 56.2 15.5 38.6 0 4 Wyoming 1001 0.58 10 99.86 6.7 14.9 31.6 0 Where: Type 1 require competency Test 0 do not require competency test Region 1-Northeast 2-Midwest 3-South 4-west Read More