Predictive Validity of English Proficiency Levels for Academic Achievement (Research Paper)

 Predictive Validity of English Proficiency Levels for Academic Achievement

   ^[1]Masoud Khalili Sabet*

^[2]Maryam Salamatbakhsh

  IJEAP- 2005-1542

Received: 2020-05-12                          Accepted: 2020-07-28                      Published: 2020-08-19

Abstract

Studies on English language teaching (ELT) in Iran suggest that it is not successful enough in producing proficient language learners. University students’ low English proficiency limits their exploitation of academic resources and raises concerns about the quality of their educations. The current study is an attempt to find out the predictive validity of English proficiency level for academic achievement in the fields of English literature and Teaching English as a Foreign Language (TEFL). To do so, a twofold procedure has been adopted. In the first phase, 83 undergraduate and postgraduate students in the above-mentioned courses of state universities in Iran were selected by convenience sampling. The correlation of their English proficiency level in three language skills of listening, reading, and writing, measured by an IELTS test, and their academic performance, measured by their current semester grade point averages (GPAs), was examined. Statistical data analysis disclosed a significantly positive correlation between the two variables under scrutiny. In the second phase, 155 students majoring in the same courses selected by convenience sampling were asked to fill out an already established 4-point Likert scale questionnaire in order to investigate the students’ perception about the effect of English proficiency on academic performance compared with other important factors. Statistical data analysis of the questionnaire revealed that ‘English proficiency’ was the eighth important factor among 35 factors influencing academic performance. The findings of the present study suggest that English proficiency functions as a cluster encompassing predictive validity for educational success as well as significant impact on students’ academic achievement. Thus, stricter measures regarding students’ English proficiency filtering should be applied for admission to the Iranian universities.

Keywords: Predictive Validity, English Proficiency, Academic Achievement, English Language Teaching

1. Introduction

As they will be presented in the next section, many studies have been carried out investigating the relationship between university students’ English proficiency level and their academic performance in general (Ghenghesh, 2015; Maleki & Zangani, 2007; Oliver, 2012). Possibly, this was recognized as one way the researchers hoped to convince the non-native students attend target language classes to promote their English proficiency level. However, few studies focused on the above-mentioned relationship in specific disciplines and courses in particular. This was recognized as a research gap in the current study as it intends to examine the function of English proficiency levels in a specific course. As far as the objectives of the previous studies are concerned, admitting international students had great financial and academic benefits for those universities. According to Zua (2016) and Reinold (2018), other than huge income earned out of international students’ tuition and their daily basis expenses, universities’ scientific and technical research could benefit largely from these students in various perspectives due to their abilities and skills flourished in different cultures and past experiences. In order to preserve such benefits and good reputation, the universities’ policy and planning agents intended to fulfill the international students’ needs particularly in the language field to maintain their higher education programs in the standard quality. In order to make sure that English proficiency level of the international students was compatible with their academic language needs, these universities used English proficiency examinations such as Test of English as a Foreign Language (TOEFL) and International English Language Testing System (IELTS) or their own domestic English proficiency tests as the gate keeper. The goal of studies about the relationship between students’ English proficiency level and their academic success in such universities was either to redesign the already existed proficiency testing system and admitting programs or to further develop English as a Second Language (ESL) programs based on the students’ language needs (Wait & Gressel, 2009).

In countries like Iran where English is not the medium of instruction in almost any university field, investigating the relationship between students’ English proficiency level and their academic performance usually follows the goal of promoting the current educational program in terms of ELT, either for the country’s educational system or for the specific university discipline by finding the areas of students’ needs in English for Academic Purposes (EAP).

 In recent decades, Iran is facing globalization like any other part of the world. According to Mitchell and Yildiz (2012), higher education is at the center of this phenomenon and economic aspect of globalization -as the most important factor- has led the Iranian universities to mass admission of new students. Currently, there are different educational divisions in Iran’s universities including day sessions, evening sessions, virtual system, Pardis section, Payam-e-Noor University, surplus capacity, etc. Students at different academic levels are admitted in the same field of study at the same university. It seems that this would lead to heterogeneity regarding students’ academic level and standard in the above-mentioned systems. Nonetheless, these students are obliged to the same curriculum and generally receive no further support from their university. The question raised here is whether this would affect the students’ academic performance, achievement and quality of higher education programs in the Iranian universities.

 The current study was an attempt to investigate the above question for EFL Iranian students. The main objective of this study was to find out whether there is a statistically significant relationship between EFL students’ English proficiency level (in three language skills of listening, reading, and writing) as indicated by their IELTS scores and their academic performance as reflected in their grade point averages (GPAs). This study also investigated how EFL students evaluate the effect of English proficiency level on their academic achievement compared with other important factors. The study is unique to some extent in the sense that such study has rarely been conducted with a sample of different educational divisions which is the reality of the current population of Iranian university students. In addition, the second phase of the study, the survey about the Iranian EFL students’ perception of the impact of English proficiency level on their academic performance, was conducted for the first time in Iran. The inquiry about the students’ perception is important as almost all significant factors affecting students’ academic achievement were taken into account. Therefore, the findings of this study would help the university authorities to revise their policy and planning regarding foreign language education programs in order to support the students whose English proficiency is not compatible with the language demands in their field of study.

2. Review of the Related Literature

Before reviewing the related literature, it is necessary to pay attention to the fact that academic achievement is affected by many factors different from one context to another, and English proficiency is just one factor among them. Therefore, the goals of such studies are quite dependent on the context of the study and its unique specifications. In the present section, some examples of studies about the relationship between academic achievement and English proficiency level in various contexts including English as a Foreign Language (EFL) and English for Specific Purposes (ESP) for various disciplines will be reviewed.

In the University of Sydney, Woodrow (2006) investigated the correlation between postgraduate students’ English proficiency indicated by their IELTS scores and their academic achievement specified by their 1ST semester GPAs in order to find out the predictive validity of IELTS examination in such particular contexts. The significant but weak correlation (R = .4, P < .01) was found between students’ overall IELTS bands and their 1st semester academic performance or GPAs. More scrutiny revealed that at the lower level of English proficiency, the correlation was stronger than at the higher level.

In another study conducted by Oliver, Vanderford and Grote, (2012) the relationship between different English language proficiency evidences and academic achievement of Non-English Speaking Background (NESB) undergraduate and postgraduate students enrolled in Curtin University in Australia over a three-year period (2006-2008) was investigated. The aim of the study was to examine the sufficiency of English language proficiency requirement for academic progression of NESB undergraduate and postgraduate students enrolling at the university. The study found that the best evidence for potential academic success refers to standardized tests, such as IELTS and TOEFL. More specifically, students submitting other documents of English language proficiency evidence tended to have more difficulties in their studies.

Tests such as TOEFL, which determine the English proficiency levels, are usually the only evaluation metric common to international applicants from a wide range of academic backgrounds for US universities branches abroad. Therefore, the result of such tests sometimes is used beyond their primary intended role to predict the applicants’ academic success. Even when the TOEFL is not used as an indicator of academic success, TOEFL score minimums are imposed as a condition for acceptance. Furthermore, the field of study may impact the degree to which English language proficiency is important to students’ academic success, so it is essential to investigate which fields of academic study are most closely tied to the measure of language ability (Barkaoui, 2017). To investigate this issue, Wait and Gressel (2009) evaluated the relationship between TOEFL and several majors in an American university in the United Arab Emirates. After statistical analyses, a positive correlation between TOEFL score and academic performance in five categories including overall GPA, engineering GPA, humanities GPA, Comprehensive Assessment Examination (CAE) pass rate, and graduation rate was identified. However, this relationship was weaker for engineering students than for students majoring in some other academic fields such as business and the arts.

Sahragard, Baharloo and Soozandehfar (2011) carried out a study in Shiraz University in which 151 undergraduate students in all academic levels majoring in English Language and Literature participated in order to find out the relationship between their English proficiency and academic achievement in this specific context. The result indicated an almost strong significantly positive relationship between students’ language proficiency level and their academic performance (r = .53, p < .05).

Maleki and Zangani (2007) also investigated the relationship between English proficiency level and academic performance of the students at Islamic Azad University of Takestan in Iran, majoring in a number of different fields. According to these researchers, one of the serious problems of Iranian university students is that many of them cannot effectively communicate in English even after graduation. They believe this issue is due to the students’ problems in General English which in turn affects their academic success. The result of the correlation between variables under the study revealed a significant relationship between English proficiency level and academic achievement (R = .48, P < .05).

The results of studies are inconsistent regarding the correlation between students’ English proficiency level and their academic performance. One reason for such a fluctuation is that academic achievement is a complex construct depending on many variables. According to Andrade (2006), Fox (2004) and Phakiti (2008), there are a wide range of non-linguistic factors that can impact the academic progress including cultural background, educational background, level of cultural adjustment, country of origin, personal characteristics, attitude, motivation, age, gender, chosen discipline, course level (undergraduate or postgraduate), level of support from social networks, personal financial issues, time available for study, ability to adapt to the (academic) culture, study strategies/practices, and class attendance (and engagement), (Andrade as cited in Oliver, Vanderford and Grote, 2012, p.4). In addition, the ability to control the above-mentioned non-linguistic factors is limited. The other reason based on Oliver et al. (2012) can be the difficulty of defining and measuring language proficiency level and academic success. Nonetheless, Oliver et al. (2012) point out that there is a general consensus on the usefulness of standardized English proficiency tests in predicting academic performance despite the existence of multiple issues regarding the predictive validity of such tests. They have recommended that individual universities and their faculties should conduct their own research to investigate the relationship between students’ academic achievements and their English proficiency level in order to set the optimum or minimum requirements for the courses they offer. So, such studies (Martirosyan, Hwang & Wanjohi, 2015) can be considered part of the evaluation programs the universities need to conduct periodically in order to maintain the quality of the higher education programs. The present study is a research attempt to seek convincing answers for the following questions:

Research Question One: Do English proficiency levels function as predictor for Iranian university students’ academic achievement?

Research Question Two: What are Iranian university students’ perceptions of the impact of English proficiency level on their academic performance?

3. Methodology

3.1. Participants

Two cohorts of students participated in the current study. The first group consisted of 83 participants who took part in an IELTS exam in order to identify their English language proficiency level. They consisted of 36 undergraduate students majoring in the English language and literature as well as 47 postgraduate students majoring in teaching English as a foreign language in the state universities of Guilan, Golestan and Mazandaran. The undergraduate students’ age ranged from 20 to 28 and the postgraduate students’ from 23-49. The participants in this group were chosen through the convenience sampling method from seven intact classes (population 217). The second cohort consisted of 155 English literature and TEFL undergraduate and postgraduate students out of 402 population in the same state universities, selected through convenience sampling method, who participated in completing a questionnaire survey in order to investigate EFL students’ perceptions about the effects of English proficiency level among other important factors on academic performance and achievement. Furthermore, two experienced IELTS instructors were employed to rate the participants’ IELTS writing tests.

3.2. Instruments and Materials

In order to measure the participants’ English proficiency, an IELTS Academic test was administered. The test was chosen from the practice test materials published by Cambridge University Press (IELTS 10). The rationale behind using IELTS Academic for this study was that according to IELTS Home Page (2019), it has a worldwide reputation on assessing English proficiency based on combination of reliable testing backed by extensive research to ensure that it meets the needs of students in higher education institutions across the world.

 In order to find out students’ perception about the effect of English proficiency on their academic performance, a questionnaire initially developed by Killen (1994) and later revised by Sibanda, Iwa, and Benedict (2015) was utilized. This questionnaire was used in the second phase of the Killen’s study on the success/failure factors influencing academic performance. Moreover, the validity of the instrument was reported in both above-mentioned studies. The questionnaire consists of thirty-five academic success factors which were presented in a Likert format including NI = not influential; SI = slightly influential; FI = fairly influential; and VI = very influential (Appendix A). The reliability of the students’ questionnaire was checked through a pilot study with EFL students. Since there were multiple Likert scale items in the questionnaire, Cronbach's alpha was used to measure the internal consistency for the questionnaire items. The reliability value of α =.895 was obtained. The reliability statistics are presented in Table 1. The reliability statistics provided the actual values for the Cronbach's alpha. Based on George and Mallery (2003) the reliability of the students’ questionnaire (α= .895) indicates that the internal consistency within the items were “good” for this specific sample.

 Table 1: Reliability Statistics of The Students’ Questionnaire

Furthermore, IELTS writing tasks band descriptors were utilized for evaluating students’ writing tasks. In addition, in order to calculate the correlation between English proficiency and academic achievement, the researcher was permitted to have access to the participants’ GPAs of their current semester as the indicator of their academic performance and achievement.

3.3. Procedures

Investigating the predictive validity of English proficiency levels for academic achievement as well as finding out the students’ perception about the effect of English proficiency on academic performance were the focal goals of the present study. To this end, a sample of EFL students in the state universities were selected from different educational divisions including day, evening, and virtual systems to participate in the study. Based on some tests, the students were different in terms of English proficiency level although they were subjected to the same academic curriculum. The procedure of gathering data for the current study was carried out in two phases. For the first phase, the IELTS examination in three sections of listening, reading and writing was given to the EFL students in seven intact classes chosen through a convenience sampling method. The test was administered towards the end of semesters in the second half of academic year 2017-2018. The rationale behind choosing the time being towards the end of the semester was to minimize the duration between the proficiency test and the end of semester examinations in order to eliminate the effect of maturation factor on students’ English proficiency. In order to rate the writing sections of the IELTS test, first the quarter of the participants’ writings were rated by three raters including the researcher. The rating was conducted using IELTS writing tasks band descriptors. In order to calculate the inter-rater reliability based on Mackey and Gass (2016), the simple percentage agreement was calculated and the agreement value of 74% among three raters was resulted. Since there was consistency among raters, the rest of the writing papers were evaluated by the researcher.

 To conduct the second phase of the present research, the survey study was carried out in some intact EFL classes. Similar to the first phase of the research, the students were given the necessary information about the study and its goals. Then they were required to complete the questionnaire indicating whether they liked to participate in the study. One hundred and fifty-five EFL students from different levels participated in this phase of the study.

4. Results

4.1. Checking the Required Assumptions

In the current study, the data were collected in terms of the students’ proficiency in three language skills of listening, reading, and writing as well as their grade point average. The data were processed and analyzed using the Statistical Package for the Social Sciences (SPSS) version 23. Before running the parametric test, the main assumptions were examined to see whether the distributions met the required assumption. As the first assumption, it was established that all the variables in the study including the IELTS scores and the GPA scores were interval data and were measured at the continuous level. The second assumption required that there was a linear relationship between the above-mentioned variables. Therefore, a scatterplot was created for each pair of variables and the two variables were plotted against each other to check for linearity. The scatter plot for the variables is presented in the following sections:

 Figure 1: Scatter Plot for The GPA Scores and Listening Scores

For the relationship between GPA scores and the listening scores, the line in the graph sloped upward, so it could be concluded that the correlation between the variables was positive.

Figure 2: Scatter Plot for The GPA Scores and Reading Scores

For the relationship between GPA scores and the reading scores, the line in the graph gradually sloped upward implying that there was a positive correlation between the two variables.

 Figure 3: Scatter Plot for The GPA Scores and Writing Scores

For the relationship between the GPA scores and the writing scores, the dots went together and formed a straight line. That meant that the two variables were related to one another. Increases or decreases in the writing scores had effect on increases or decreases in GPA scores. For the third assumption of the Pearson test, the outliers were examined through the explore command. Since Pearson's r is sensitive to outliers, the distribution of scores was examined in terms of the outliers. There were no significant outliers in the distributions and the data points followed the usual pattern. This was examined through computing the trimmed means for the variables and comparing them against the original means. Finally, based on the fourth assumption, the variables were assessed to see if they were approximatelynormally distributed. In order to assess the normality of the distributions, skewness analysis was used. The results are presented in Table 2.

Table 2: Normality Assumption for The Distributions

In Table 2, the descriptive statistics and other information concerning the variables were provided. This information included trimmed mean, skewness, kurtosis, standard deviation, and 95% Confidence Interval for Mean. To obtain the 5% Trimmed Mean, first, the top and bottom 5 percent of the cases were removed and a new mean value was calculated. Next, the original means and the new trimmed means were compared for all the tests, and it was found that extreme scores did not have a strong influence on the means. In other words, since the trimmed means and mean values were not very different and the two mean values were very similar for the IELTS scores and the GPA scores, the values were not too different from the remaining distribution and thus these cases were retained in the data file. In addition, the Skewness and kurtosis values that were within the range of (+2) were also provided as part of this output that gave information about the normal distribution of scores in all tests. After examining the main assumptions for the Pearson test, the main statistical analyses were run to provide the answers to the research questions.

4.2. Investigating the Research Questions

The Pearson correlation coefficient was used to examine the strength and direction of the linear relationship between the two continuous variables including the students’ IELTS scores in three language skills of listening, reading, and writing and their GPA scores.

Table 3: Pearson Correlation Test for The Students’ Listening Scores and The GPA Scores

**Correlation Is Significant at the 0.01 Level (2-Tailed).

The p-values were compared to the significance level. In Table 3, the p-value for the correlation between listening scores and GPA is less than the significance level of 0.01, which indicates that the correlation coefficient is significant (r (83) = .297, p < .01). Based on the statistical guidelines (Evan, 1996; Mackey and Gass, 2016) this relationship is weak. Therefore, the null hypothesis is rejected implying that there is significant relationship between students’ listening proficiency (IETLS Listening Score) and their academic performance (GPAs).

Table 4: Pearson Correlation Test for The Students’ Reading Score and The GPA Scores

 *Correlation Is Significant at the 0.05 Level (2-Tailed).

As it is shown in Table 4, the Pearson Correlation came to .233 with a p value of .034. Since the p value is lower than the alpha level (0.05), it was concluded that there is meaningful linear relationship between the learners’ reading proficiency and their GPA (r (83) = .233, p < .05). Based on Mackey and Gass (2016) guidelines, this relationship is weak. Consequently, the null hypothesis is rejected implying that there is statistically significant relationship between students’ reading proficiency (IETLS reading score) and their academic performance (GPAs). To examine the relationship between the learners’ writing proficiency and their GPA, another Pearson’s Product Moment Correlation analysis was run.

Table 5: Pearson Correlation Test for The Students’ Writing Scores and The GPA Scores

 ** Correlation is Significant at the 0.01 Level (2-Tailed).

The results of Pearson correlation coefficients revealed a statistically significant positive relationship between the learners’ writing proficiency and their academic performance (r (83) = .358, p < .01). According to Mackey and Gass (2016) suggestion, this relationship is almost moderate. So, the null hypothesis is rejected suggesting that there is statistically significant relationship between students’ English writing proficiency and their academic performance.

The actual size of Pearson’s (r) and the amount of shared variance between the two variables were also obtained. It was calculated by squaring the value of the correlation coefficient. The shared variances between the variables are in Table 4.5. As it is displayed in table 4.5, IELTS writing scores were better predictors of GPA scores compared to listening and reading scores.

Table 6: Shared Variances Among the Variables

In order to find out the most significant factors (top ten) affecting students’ academic performance based on their own perception, item statistics were computed for the items of the questionnaire (Table 4.6). The highest mean ranks were reported for item 18 “self-motivation” and item four “hardworking, commitment and dedication” for which many respondents rated VI (very influential) as the most significant factors that affected their academic performance. The respondents also reported high mean rank for “self-confidence”, “Ability to understand in-depth content”, “Self-discipline”, “regular study”, Ability to work independently as well as “English proficiency”. “Efficient time management and organizational skills” and “Having good writing skills” were the next two effective factors in developing the students’ academic performance based on their own perception.

 Table 7: Item Statistics for the Questionnaire

5. Discussion and Conclusion

The current study has been conducted in two phases. In the first place, it was intended to investigate the possible predictive validity of Iranian EFL students’ English proficiency levels in three language skills of listening, reading, and writing on their academic achievement. Although the correlation analysis indicated that there was a significant positive relationship between the students’ English proficiency levels in three mentioned language skills and their academic achievements, the strength of the correlation between the variables was weak according to Mackey and Gass, (2016) guidelines. This would possibly suggest that English proficiency in three language skills of listening, reading and writing are not strong predictors of how EFL students would perform academically in Iranian universities. In the second phase In addition, the researchers intended to investigate the students’ perceptions of the impact of English proficiency levels on their academic performance. The outcome of the questionnaire survey can help to interpret such results to a great extent. In this survey, 155 EFL students were asked to evaluate 35 factors affecting their academic performance. Statistical analysis of the survey data indicates that the top ten factors affecting EFL students’ academic performance are: 1) self-motivation, 2) hardworking, commitment and dedication, 3) self-confidence, 4) Ability to understand in-depth content, 5) Self-discipline, 6) regular study, 7) Ability to work independently, 8) English proficiency, 9) Efficient time management and organizational skills, and 10) Having good writing skills respectively. Based on the students’ perception, there are seven factors which are more important than English proficiency justifying to some extent the weak correlation between the variables in this study.

By more scrutinizing in the results of the study, it is understood that correlation coefficient for the writing part (r = 0.358, p<0.01) is bigger compared to those for the listening (r = 0.297, p < 0.01) and reading (r = 0.233, p <0.05) sections of the IELTS test. It means that proficiency in the writing skill is more effective for gaining better academic results. On the other hand reading scores and GPA indicate the weakest correlation. The possible explanation for this could be the evaluations and assignments in Iranian universities which are mostly in written form. The weak correlation between variables can also be interpreted as the result of the nature of the proficiency test. Since the test was not a high stake one, this may have influenced the students’ performance meaning that they did not do their best on the test. Another interpretation for such finding can be the different strategies students use in order to cope with the language demands of their field of study. In addition, the lecturers may also compromise the way they deliver their course or evaluate students’ assignments in order to accommodate their students’ language abilities. For example, when grading students’ writings, they may ignore the grammar mistakes and focus on the content or they may have to adjust the depth and difficulty of content covered in their class or to alter their method of delivery. More precise interpretation for findings of this study demands more investigation in a separate study which is out of the scope of the current study.

 The findings of the current research are compatible with some other studies conducted worldwide. More specifically, studies carried out by Green (2014); Harsch, Ushioda, and Ladroue, (2017); Morris and Maxey (2014); Kamrul Hasan and Moniruzzaman Akhand (2014); Sadeghi, Moshtaghi Kashani, Maleki and Haghdoost (2013); Martirousyian, Hwng and Wanjohi (2015); Woodrow (2006); and Zua (2016)  have all come up with similar results. In fact, all of them found a significant correlation between English proficiency and academic achievement. The reason can be either such studies were conducted on a sample consisting of graduate or postgraduate students whose English proficiency usually played more important role for academic success, or the sample consisting of students whose major study was closely affected by English proficiency such as EFL, MBA, Social sciences and Education courses.  On the other hand, there are some studies which their findings contradicted the findings of the current study. For instance, studies carried out by Chen and Sun (2006), Fass-Holmes and Vaughn (2015), and Fox (2004) found no correlation between university students’ English proficiency and their academic performance. The justification for such discrepancy might be the fact that the participants’ major study which was less affected by the students’ English proficiency levels. To render some examples we can mention the courses in engineering field where some context-related factors other than English proficiency played relatively more important roles for academic achievement. More specifically, some language-irrelevant variables like technical construct could possibly overshadow the role of English proficiency on the subjects’ academic performance.

It is worth mentioning that the findings of this study were affected by some limitations. The first limitation refers to the participants who were made aware of the aim of the IELTS test prior to taking it and as it was not a high-stake examination, some participants might not put enough effort to get the best result. Secondly, this study would come to the best result if all the participants were chosen from the same academic level since this would equalize the courses taken by the participants for calculating GPA. The third limitation refers to the number of students participating in this study which despite the researcher’s exhaustive effort could not exceed 83. Another limitation of the study was that out of 83 participants, only 12 students agreed to take part in the speaking section of the test. Therefore, the speaking section was excluded from the proficiency test. Finally, there are many other important factors influencing academic success among which English proficiency is just one factor. It is impossible to control all other factors affecting academic achievement in order to reach more reliable results.

Despite the limitations, this study has important implications for foreign education policy and planning actors in different levels from university faculty members to authorities in the ministry of education. As the descriptive statistics in table 4.4 suggest, English proficiency level of EFL students in Iran generally is below the norm accepted in the international universities for ESL/EFL courses. In order to raise the level of students’ English proficiency at universities, the authorities should either use a standard proficiency test such as TOFLE or IELTS with minimum entry level as a gate keeper or consider holding make up programs for the students whose English proficiency level does not meet the minimum requirements.

Another implication of the current study for the faculty members and the policy makers in universities is to pay due attention and take effective actions in order to improve the students’ academic writing skills as the finding of the study clearly indicates its effective role in academic success of the students. At the ministry level, the authorities should consider a multifaceted reform program for the English education in the public sector. A program which not only consists of textbooks and curriculum reforms but also considers other important aspects for implementation of a successful language learning program such as training qualified teachers, considering the context and the students’ needs in choosing the ELT methodology, allocating fair amount of time to the language programs, providing necessary financial and technological supports, encouraging and educating the students about the importance of learning English, etc. The current study has also some implications for the university EFL students in both undergraduate and postgraduate levels as it gives them insight about the English proficiency level requirements in order to be more successful academically.

 Finally, as it was already stated the limitations of the current study had some negative effects on the results achieved from this research. Providing the future researchers with the chance to overcome the obstacles causing limitations for this research, they will definitely achieve more precise results. For example if all the participants are chosen from the same academic level, studying the same courses taught by the same lecturers, the study will definitely give more reliable results. Future research can also include the investigation about the effects of other important context related variables on academic success along with English proficiency such as the professors’ attitudes towards their students’ English proficiency, the strategies students use to manage the language demands of the courses delivering in English, students’ main goals of higher education, etc. It would also be helpful to investigate the correlation between English proficiency and academic success for other disciplines, especially for science and technology related fields in the postgraduate levels and upper. In addition the future research can be conducted considering a greater number of participants (more than a hundred) in order to increase the accuracy of the findings. Furthermore such research can be conducted in different universities, since each university has its unique contextual conditions that have their own effects on the students’ academic success.

6. References

Barkaoui, K. (2017). Examining repeaters’ performance on second language proficiency tests: A review and a call for research. Language Assessment Quarterly, 14(4), 420-431.

Chen, Y., & Sun, C. (2006). Language Proficiency and Academic Performance, Proceedings of the 11th Conference of Pan-Pacific Association of Applied Linguistics, Korea, 11, 58–72.

Evans, J. D. (1996). Straightforward statistics for the behavioral science. Pacific Grove, CA: Books/Cole Publishing.

Fass-Holmes, B. & Vaughn, A. A. (2015). Evidence that international undergraduates can succeed academically despite struggling with English, Journal of International Students, 5(3), 228-243.

Fox, J. (2004). Test decisions over time: Tracking validity. Language Testing, 21(4), 437–465. doi:10.1191/0265532204lt292oa.

George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and reference, 11.0 update. 4th Edition, Boston: Allyn & Bacon.

Ghenghesh, P. (2015). The relationship between English language proficiency and academic performance of university students – Should academic institutions really be concerned? International Journal of Applied Linguistics & English Literature, 4(2), 91-97.

Green, J. A. (2014). The effect of English proficiency and ethnicity on academic performance and progress. Advances in Health Sciences Education, 20(1), 219–228.

Harsch, C., Ushioda, E., & Ladroue, C. (2017). Investigating the predictive validity of TOEFL IBT test scores and their use in informing policy in a United Kingdom university setting. ETS Research Report Series, 2017(1), 1–80. doi:10.1002/ets2.12167.

IELTS Home page. (2019). Retrieved March 30, 2019 from https://www.ielts.org/news/2019/ielts grows-to-three-and-a-half-million-a-year.

IELTS Writing Tasks Band Descriptors. Retrieved Mackey, A & Gass, S. (2016). Second language research, methodology and design. New York, NY: Routledge.June 24, 2018 from   https://idpielts.me/wp-content/uploads/2017/08/Writing-Band-descriptors-Task-1-and-2_Public pdf.

Kamrul Hasan, M. & Moniruzzaman Akhand, M. (2014). EFL students’ English language proficiency and their academic achievement. ABAC Journal, 34(2), 64-70.

Killen, R. (1994), Differences between students' and lecturers' perceptions of factors influencing students' academic success at university. Higher Education Research & Development, 13(2), 199-211.

Maleki, A., & Zangani, E. (2007). A survey on the relationship between English language proficiency and the academic achievement of Iranian EFL students, The Asian EFL Journal Quarterly, 9(1), 65-85.

Martirosyan, N., Hwang, E., & Wanjohi, R. (2015). Impact of English proficiency on academic performance of international students. Journal of International Students, 5(1), 60-71.

Mitchell, D. E. & Yildiz Nielsen, S. (2012). Internationalization and Globalization in Higher Education, In H. Cuadra-Montiel (Es.), Globalization - Education and Management Agendas. (pp.4-22).

Morris, M., & Maxey, S. (2014). The importance of English language competency in the academic success of international accounting students. Journal of Education for Business, 89(4), 178–185.

Oliver, R., Vanderford, S. & Grote, E. (2012). The relationship between types of evidence of English language proficiency and academic achievement of non-English speaking background students at an Australian university, Retrieved April 17, 2018 from https://espace.curtin.edu.au.

Reinold, J. (2018). The benefits of the internationalization of higher education, Retrieved February 8, 2019, from https://www.maastrichtuniversity.nl/blog/

Pallant, J. (2010). SPSS survival manual: Step by step guide to data analysis using SPSS Program. 4^th Edition, New York: McGraw Hill.

Sadeghi, B., Moshtaghi Kashani, N., Maleki, A., & Haghdoost, A. (2013). English language proficiency as a predictor of academic achievement among medical students in Iran, Theory and Practice in Language Studies, 3(12), 2315-2321.

Sahragard, R., Baharloo, A., & Soozandehfar, S. M. A. (2011). A closer look at the relationship between academic achievement and language proficiency among Iranian EFL students, Theory and Practice in Language Studies, 1(12), 1740-1748. doi:10.4304/tpls.1.12.1740-1748.

Sibanda, L., Iwu, C. G., and Benedict, O. H. (2015). Factors influencing academic performance of university students, Demography and Social Economy 2 (24), 103-115.

Wait, I. W. and Gressel, J. W. (2009). Relationship between TOEFL score and academic success for international engineering students, Journal of engineering education, 98(4), 389-398.

Woodrow, L. (2006). Academic success of international postgraduate education students and the role of English proficiency, University of Sydney Papers in TESOL, 1, 51-70.

Zua, B. (2016). The impact of language proficiency on the academic achievement of international students, International Journal of Education and Human Developments, 2(6), 33-43.

Appendix A: Questionnaire on academic success factors

This questionnaire forms the basis of the research methodology. Please note that the response will be treated with confidentiality and will be used for research purpose only.

How do the below factors influence your academic performance? Mark ‘X’ in the appropriate boxes, where: NI = Not Influential; SI = Slightly Influential; FI = Fairly Influential; VI = Very Influential