International Journal of Science Commerce and Humanities Volume No 2 No 7 October 2014

The internationalization of higher education in U.S. Colleges and Universities has raised issues concerning the performance of international students. Most of them speak English as a foreign language; therefore, this study is intended to analyze to which degree the level of English proficiency of an international student impacts his or her academic performance. This research is focused on the international undergraduate student body in an AACSB Accredited College of Business Administration, in which a sample of 34 students from various nations was used. The tool used to measure academic performance was the Grade Point Average system (GPA), which ranges from 0 (minimum) to 4.0 (maximum). Moreover, English proficiency was measured by the TOEFL iBT® (Test of English as a Foreign Language Internet Based), scored on a scale from 0 (minimum) to 120 (maximum) points. The decision to use the TOEFL as an indicator was based on the fact that the exam tests one’s ability to understand English in an academic environment. In addition to comparing overall academic performance related to English proficiency, as control variables, we tested further to determine if age, ethnicity, major, and gender are determinant in explaining the performance of an international undergraduate student in an English speaking academic environment. Furthermore, we also compared the academic performance of the international students by adding to the sample 39 American students in order to analyze any existing correlation. The aspiration of this study is to find enough evidence to support the tentative conclusion that the higher the level of English proficiency, the less important this factor is in determining academic success. In other words, if a student shows to be proficient and does not succeed academically, other factors can be attributed to his or her lack of success. Introduction In the United States, admitting students from various nationalities of the world to study in American universities has internationalized the undergraduate education. The large number of international students who attend an American university can be attributed to the quality of the education, infrastructure, and attractiveness of the language among others factors. Nevertheless, this raises the issue to whether learning in a foreign language, for those who are not native English speakers, has any impact on academic performance. This article reviews research in this field and presents a study of 34 international students at the College of Commerce and Business Administration at Jacksonville State University. International Journal of Science Commerce and Humanities Volume No 2 No 7 October 2014 84 Jacksonville State University is a public, regional school that currently has about 8700 total enrollment, with 7650 undergraduates. The College of Commerce and Business Administration constitutes about 950 students. The student population is diverse, with about 63% white, 28% black and the rest coming from Hispanic, Asian, or Native American heritage. Women make up 58% of total enrollment, whereas men represent 42%. The international student body of the university represents about 4% of total enrollment. Although it seems plausible to assume that proficiency in English is fundamental for academic success, it is necessary to point out that communication abilities are more important in some areas (e.g., English literature) than in others (e.g., mathematics). Therefore, despite the correlation between TOEFL scores and academic performance, the TOEFL score may be misused if looked upon as the only indicator of how well a student might perform [Wait and Gressel, 2009]. For this reason, this paper extends the research to determine if different majors within the business course (e.g., economics, finance, accounting, marketing, management), as well as gender, age, and ethnicity are factors in explaining the relationship between English proficiency and academic performance in an undergraduate business program. Additionally, the research also covers a comparison between international students and American students. Literature Review Wait and Gressel [2009] evaluated relationships between TOEFL scores and several measures of academic success of students at an American university abroad. Characterizing these relationships helps assess the scope of the TOEFL score’s use in admissions decision making. Their conclusion was that in spite of correlations between TOEFL scores and academic performance, TOEFL scores should not be used beyond assessing individual student’s English proficiency. Cho and Bridgeman [2012] used a similar model to examine the relationship between scores on the TOEFL Internet-Based Test (TOEFLiBT®) and academic performance in higher education, defined in terms of grade point average (GPA). They used correlation-based analyses for subgroups by academic status and disciplines. Although the predictive validity expressed in terms of correlation did not appear to be strong, the general pattern shown indicated that students with higher TOEFLiBT scores tended to earn higher GPAs and that the TOEFLiBT provided information about the future academic performance of non-native English speaking students beyond that provided by other admissions tests. Such observations led them to conclude that even a small correlation might indicate a meaningful relationship between TOEFLiBT scores and GPA. Nelson, Nelson, and Malone [2004] analyzed the retention and completion rates of international students seeking a master's degree at an American university. Predictor variables analyzed to determine retention to degree completion included TOEFL scores (both raw and composite), age, gender, geographic categories of native country, native language, academic area of concentration (grouped into five general areas of study), graduate grade point average in the first nine hours of graduate study, and admission status (regular or probationary). The objective of the study was to find those criteria that had predictability to help admissions officers screen applicants for graduate study. This study provided helpful insight on which factors can be attributed to the success of an international student. Our study expands on their use of TOEFL scores, age, and gender. Lun, Fischer, and Ward [2010] examined explanations for the observed differences in critical thinking between Asian and New Zealand (NZ) European students, and tested hypotheses derived from research in international education and cultural psychology. The results showed that NZ European students performed better on two objective measures of critical thinking skills than Asian students. English proficiency, but not dialectical thinking style, could at least partially if not fully explain these differences. The results also indicated International Journal of Science Commerce and Humanities Volume No 2 No 7 October 2014 85 that Asian students tended to rely more on dialectical thinking to solve critical thinking problems than their Western counterparts. They were able to conclude that there are significant differences in a student’s critical thinking abilities according to ethnicity. Methodology and Results Analysis of a student’s academic performance was measured by the grade point average system (GPA) as the dependent variable of the model using the least squares regression method. The independent variables chosen were: TOEFL score, ethnicity, gender, age, and major. To measure ethnicity with a relatively small sample of 34 students, we divided the sample into non-minority students, who are white, and the minority group that included Africans, Hispanics, and Asians. The majors were broken down into Accounting, Finance & Economics, and Management & Marketing. Finance & Economics and Management & Marketing were grouped together due to the similar characteristics of each major. Descriptive statistics for the relationship between GPA as the dependent variable and TOEFL score as the independent variable are given in Table 1. Table 2 adds other independent variables such as age, ethnicity, and gender to show their relationship with the dependent variable. Tables 3, 4, and 5 show the regression broken down into majors and the differences between them. Finally, Table 6 adds to the original sample of 39 American students for comparison purposes. Table 1 Regression of GPA on the TOEFL Scores Included Observations: 34 Coefficient Std. Error P-value TOEFL 0.017 0.006 0.007 As expected, the regression shows that TOEFL scores have a positive coefficient related to the dependent variable GPA. Wait and Gressel [2009], Cho and Bridgeman [2012], and Nelson, Nelson, and Malone [2004] also showed in their work that a positive relationship exists between these variables. However, in our sample this coefficient was relatively small with a value of 0.016952, but it was significant at the 0.007 level.