THE ROLE OF ESP COURSES IN GENERAL ENGLISH PROFICIENCY

The purpose of the present paper is the study of the interaction between learning English for Specific Purposes (ESP), in particular, English for the Financial Sector, and general English proficiency. The research examines the effects of an ESP course being taught for a year on the students' general English proficiency.Two sets of tests were prepared for that purpose and administered to 30 first-year students of finance and law. The students took the placement test twice, at the beginning and at the end of the school year. To monitor test performance over a research period, a parallel form measuring the same competences was administered at the beginning of the second semester. In the test development process a special consideration has been paid to the level of difficulty and its relation to the students' prior educational context. Drawing on the National State Matura exams the test is set at Common European Framework of Reference for Languages (CEFR) Level B2. As regards its content the test is comprised of reading comprehension tasks (multiple matching, multiple-choice cloze, gapped text) and grammar tasks aiming to examine lexical and grammatical competence.There were two major assumptions in this study: 1) Learning ESP can improve students' general English proficiency, and 2) There is a more substantial improvement in lexical competence as compared to the improvement in grammatical competence.There is strong evidence in support of the first hypothesis, whereas for the second one the results were ambiguous. After major findings are presented and discussed, implications for ESP teaching are given in closing.


INTRODUCTION
English teaching and learning has a very long tradition at the global level.In recent years rapid developments in English teaching and its unprecedented global spread have been fostered through its unique function as the world language of both business and communication and its role as lingua franca.The introductory part of this paper points to the specificities that differentiate teaching English for General Purposes (EGP) and English for Specific Purposes (ESP).Taking into consideration a variety of approaches and definitions of ESP (as given by Carter 1983;Dudley-Evans/St. John 1998;Harding 2007;Hutchinson/Waters 1987;Mackay/Mountford 1978) several characteristics have been chosen to indicate similarities and differences between EGP and ESP.A brief overview is given in Table 1.
One of the differences between EGP and ESP lies in the learners and their purposes for learning English.EGP is usually taught at earlier stages of language learning to primary and secondary school students who have some knowledge of English or language system and their proficiency level is rather low or pre-intermediate.Most ESP courses are taught at later stages of language learning on students at higher education institutions and to adult learners, but due to its methodology ESP can also be used with beginners (Dudley-Evans/St.John 1998).ESP students are usually students who already have some knowledge of or experience with English.Research shows (Martinović/Poljaković 2010;Jelovčić 2008) that groups of first year university students exhibit significant inhomogeneity in terms of proficiency level and are comprised of students whose knowledge of English is at the advanced level but also those who have lower language proficiency as described by B2 "can do" descriptors of CEFR (Council of Europe).Characterised as being learner-oriented, the teaching process in an ESP course differs from generic foreign language teaching in that it is needs oriented and focused on the studentʼs use of the English language in academic and professional contexts.ESP aims at developing language proficiency in response to the specific needs of learners in future occupational settings.In answering the question Why does the learner need to learn a foreign language?Hutchinson and Waters (1987) pointed to the notion of target situation needs and learning needs and that awareness of a need will have an influence on what will be accepted as reasonable content in the language course.Therefore, although it is hard to predict general English learnersʼ future needs, there is always a perceived and identifiable need of some sort (Hutchinson/Waters 1987).EGP provides a broad foundation and development of four language skills rather than a detailed, a selective specification of goals like ESP.
Another ESP distinguishing feature is its specific professional or scientific terminology.The specific technical terminology is used under certain conditions and by a specific group of people belonging to a particular profession.Learners who are fluent in English may still have a lack of knowledge of specific vocabulary used in the specific vocational field.
In an EGP course an emphasis is placed on the language as a system, and situations that are used for teaching and learning are of common or general nature.An ESP course combines English language teaching and vocational content.This combination has a pragmatic value and is highly motivating because students are able to apply and integrate their professional knowledge (content knowledge in their main field of study) and what they learn in English classes (Cigan 2013).
Harding indicates that in ESP practical application and use of language overrides other aspects of language learning (Harding 2007).ESP concentrates more on language in context than on teaching grammar and language structures.Hutchinson and Waters found that the teaching of grammar is based on the minimum necessary for understanding academic and technical texts.Dudley-Evans and St. John appear to agree that it is incorrect to consider grammar teaching as outside the remit of ESP.How much priority is given to teaching grammatical forms depends on the learnerʼs level in English and whether learnerʼs weaknesses interfere with comprehension of meaning and linguistic production in accurate English (Dudley-Evans/St.John 1998).
In the present study the construct of lexical and grammatical competence is viewed in light of communicative language competence as defined in the Common European Framework of Reference for Languages (CEFR), in Richardsʼs, Mearaʼs and Canale and Swainʼs taxonomies.In the CEFR, the lexical competence is defined as "the knowledge of and ability to use the vocabulary of a language" consisting of lexical elements and grammatical elements.Illustrative scales are listed in respect of vocabulary range and the ability to control this knowledge."Grammatical competence may be defined as knowledge of and ability to use the grammatical resources of a language.It is the ability to understand and express meaning by producing and recognising well-formed phrases and sentences in accordance with these principles" (Council of Europe 110-112).These descriptions can be reviewed and evaluated from different perspectives of the literature and research on these notions, but that is not within the focus of this study.
In an attempt to explicate further these two concepts, we focus on the question of what it means to know a word.Richards (1976) introduced a specification of seven main aspects of word knowledge: a) knowing the degree of probability of encountering a word in speech or print, b) knowing the limitations imposed on the use of the word according to function and situation, c) knowing the syntactic behaviour associated with the word, d) knowing the underlying form of the word and the derivatives that can be made of it, e) knowing the associations between the word and other words in the language, f) knowing the semantic value of the word, and g) knowing many of the different meanings associated with the word.Nation (2001) extended his list to eight types of word knowledge and pointed to receptive and productive knowledge.He proposed the following aspects of lexical knowledge: a) form (spoken and written form), b) position (grammatical behaviour and collocational patterns), c) function (word frequency and appropriateness), d) meaning (conceptual content and word associations).
One of Mearaʼs (1996) crucial ideas is focusing not just on individual words, but on a small number of global properties.Lexical competence is probably not just the sum of the speakerʼs knowledge of the individual items their lexicons contain.Lexical competence is measured by both the size of a learnerʼs storage of lexical items and the organization of those items.Canale and Swain (1980) understand grammatical competence as a type of competence that includes knowledge of lexical items and of the rules of morphology, syntax, sentence-grammar semantics, and phonology.Grammatical competence is the ability to recognize and produce the distinctive grammatical structures of a language and to use them effectively in communication.This research examines both grammatical and lexical competence in view of the development of foreign language communicative skills in a professional context.

Aim of the Study
The aim of the study was to determine whether there is an interactive relationship between learning English for Specific Purposes, in particular English for the Financial Sector at the Effectus College in Zagreb, and general English proficiency.This research explores the effects of the ESP course being taught for a year on the studentʼs general English proficiency by examining studentʼs performances on three tests over a one-year period (during the 2012-2013 academic year).
Two major assumptions were made and examined: 1) Learning ESP can improve studentsʼ general English proficiency.
2) There is a more substantial improvement in lexical competence as compared to the improvement in grammatical competence.
All figures and tables presented in this paper have been created using the data gathered within the research framework.

Participants
A total of 30 first-year students at the Effectus College in Zagreb, enrolled in the study programme of finance and law, both male (N = 15) and female (N = 15), participated in the study.Participants were full-time and part-time students aged 19 and older.11 participants in this sample are part-time students and nine of them are currently employed.The participants enrolled the 2012-2013 academic year after completing secondary education of either a four-year vocational school (N = 17), a grammar school (N = 11) and a three-year vocational school (N = 2).
Data on studentsʼ previous experience regarding foreign language learning were also collected on the basis of studentsʼ self-report about languages learned, duration of learning and grades obtained.Twenty-eight students indicated that they had learned English.Seven of them (23%) have learned English for less than four years.Twelve students (40%) have learned English for eight to nine years, whereas only four students (13%) have learned English for 12 years.Nearly one third of the students (N = 9) reported having obtained a grade of Very Good, eight students (26%) a grade of Good, seven students (23%) a grade of Sufficient, whereas only four students (13%) reported to have achieved a grade of Excellent in English language in the last year of secondary education.[The Croatian national grading system consists of five grades: Excellent (5) (highest grade), Very Good (4), Good (3), Sufficient (2) (minimum pass grade), and Insufficient (1) (fail)].In this sample 13 students (43%) indicated that they had learned German.Six of these students learned German for four years.Five students learned Italian and only one student learned French.

Instruments
The primary instruments used in this study were two general English proficiency tests and a questionnaire with demographic variables and a set of questions about prior foreign language learning attached to the first test.General English proficiency tests had been designed on the basis of the National State Matura exams (higher level) paying special consideration to the level of difficulty and studentsʼ prior educational context.Based on the National Curriculum Framework (NCF, 2010) which determines expected studentʼs achievements at the completion of secondary school education and draws on CEFR levels, the tests were set at CEFR B2 level.The tests comprised two basic parts.The first part, reading comprehension assessing lexical competence, consisted of four tasks, which involved multiple-choice items on a text with four options, multiple matching, multiplechoice cloze and a gapped text (sentence reordering).In terms of types of texts used, texts assessing reading comprehension were taken from newspapers, magazines, journals and encyclopaedia and used in the National State Matura exam of higher level (June 2012).The second part, assessing grammatical competence, consisted of three tasks.Two tasks involved multiple-choice items and consisted of a sentence in which out of four options offered only one was a correct answer.The last task was a gapped sentence -open cloze and students were required to find the correct word for each gap.The tests comprise 52 items each.The key features of these tasks are shown in Table 2.
The questionnaire comprised two parts: a set of independent variables, including the basic demographic characteristics of respondents (gender, student status in terms of employment, parentsʼ education) and the second part comprising questions to obtain the background information concerning participantsʼ secondary education completed, foreign languages learned and grades achieved.

Procedure
The respondents took the first test at the beginning of the first semester (October 2012) and at the end of the second semester (June 2013).To monitor test performance and assess progress over the research period, a parallel form measuring the same competences was administered at the beginning of the second semester (March 2013).In terms of level of difficulty and types of tasks comprised, the parallel form corresponded to the first, i.e. last test to the full.Over the research period (2012-2013 academic year) students were taught a total of 90 hours of lessons in ESP.
All three tests for determining general English proficiency were taken in controlled classroom settings.The tests were administered to students during their regular classes.The administration time for every test was 60 minutes.The students were informed that the tests were provided with codes based on which only the teacher/principal researcher would know the identity of the respondents.

Descriptive statistics
In this paper the following codes and abbreviations will be used while reporting on data analysis and results: Test 1, also T1 = general English proficiency test administered at the beginning of the first semester Lexis 1, also L1 = refers to tasks 1-4 of Test 1 Grammar 1, also G1 = refers to tasks 5-7 of Test 1 PF = refers to the parallel form test administered at the beginning of the second semester Lexis PF, also LPF = refers to tasks 1-4 in the parallel form Grammar PF, also GPF = refers to tasks 5-7 in the parallel form Test 2, also T2 = test administered at the end of the second semester (Test 1 administered for the second time) Lexis 2, also L2 = refers to tasks 1-4 in Test 2 Grammar 2, also G2 = refers to tasks 5-7 in Test 2 The first phase of data analysis encompassed computing achieved scores in all three tests.Means and standard deviations are presented in Table 3.The total number of scores for each test is 52.The fi rst part of the test examining lexical competence consists of 30 points, whereas the second part of the test examining grammatical competence consists of 22 points.A comparison of the scores participants obtained on Test 1 and Test 2 show that they performed better on Test 2 (Mean difference = +1.63).However, comparing scores for L1, L2 and G1 and G2 we notice a very small, almost insignifi cant decline in scores on L2 (Mean difference = -0.03)and a small increase on G2 (Mean difference = +1.66).A comparison of the scores indicates that the maximum score obtained on all three tests is 43 points.The minimum score obtained on Test 1 is 6 points; on Parallel form 7 points and the minimum score on Test 2 is 10 points (the highest among these three).The maximum score in grammatical tasks is almost the same in all three tests: G1 = 19 points, GPF and G2 = 20 points, while the score for lexical tasks is the same in the fi rst two tests (25 points) and decreases somewhat on the last one (23 points).These scores are also presented in Figure 3.The mean score rankings of average percentage of items answered correctly for two test sections and three test administrations (all students on all test items) are shown in Table 4. Average performance in grammatical tasks of the Parallel Form is ranked fi rst (53.5%) followed by the performance in grammatical tasks of the Test 2 (53.3%).Performance in lexical tasks of the Parallel form with 47.8% is ranked in the middle together with performance in grammatical tasks in Test 1 (45.8%).Performance in lexical tasks in the Test 1 and Test 2 are ranked low.Students performed better on the Parallel form than on Test 1 and/or Test 2 in both lexical and grammatical tasks (Figure 4).The reason for better performance can be found, amongst other possible influences, in the time at which the test was taken.Students took the Parallel form at the beginning of the second semester after they had passed exams in other courses so that they felt more relaxed and were able to concentrate better than when they took Test 2 at the end of academic year.To assess whether there is a statistically significant difference between the scores in particular test sections and total test scores, a T-test for paired samples was conducted.It revealed that there is a statistically significant mean score gain from the lexical test Another statistically significant mean score difference has been found between the lexical section of the Parallel form and the lexical section of Test 2 (t (26) = 3.392, p = 0.002) and between the total scores of Test 1 and the total score of the Parallel form (t (26) = -3.045,p = 0.005).
In this paper, differences between respective test scores per student were also examined.Frequencies of studentsʼ answers reveal that only five students (18.5%) performed worse on the Parallel form and 22 students (81.5%) performed better in comparison to their scores obtained on Test 1.When we compared scores obtained on the Parallel Form and Test 2, one student obtained equal score on both tests, whereas 17 students (63%) performed worse and nine students (33.3%) performed better.The analysis of results for the Test 1 and Test 2 showed that one student obtained an equal score, 11 students (36.7%) performed worse on Test 2 and 18 students (60%) performed better on the Test 2. The comparison of the test overall scores of T1 and PF and T1 and T2 reveal that the average total scores show an upward trend, which supports the first hypothesis.
Changes in the total scores of the proficiency tests taken suggest that students scored higher after learning ESP in particular, they scored substantially better after the first semester (+five points on PF) and moderately better after the second semester (+two points).This positive progress trend has been demonstrated better on the Parallel form when students were not as worried about passing exams in other courses as was the case at the end of the second semester.

Relations among tests scores and ESP grades
To examine the relationship between the grades obtained in ESP taught over the research period and performance on the general English proficiency tests (T1 and T2) the Pearson Correlation was run.The correlations coefficients are presented in Table 5.
Due to a rather small number of participants (N = 30), we considered only correlations at the 0.01 significance level.For the purpose of this analysis, grades were used as single values: ESP 1 (grades obtained at the end of the first semester), ESP 2 (grades obtained at the end of the second semester), Test 1 grade and Test 2 grade.A statistically significant and good (in terms of strength) relationship has been found between the grade in ESP 2 and the performance on Test 1 (r = 0.622, p < 0.01).Correlation analysis revealed a high correlation between the grades obtained on the Test 1 and Test 2 (r = 0.793, p < 0.01).Likewise, there is a good correlation between grades in ESP 1 and ESP 2 (r = 0.667, p < 0.01).
It is interesting to note that the grade in ESP 2 correlates with T1 and T2 results much better than ESP 1. Also, correlations are almost by double higher between ESP 2 and Test 1 than between ESP 1 and Test 1.Thus, the grade at the end of the academic year seems to be a better indicator of studentsʼ overall achievement.Those correlations between grades in ESP and studentsʼ scores on additional research tests indicate association but at this point do not provide information on cause-and-effect relation, because there can be many other factors that have an effect on the studentsʼ performance.In order to examine the impact of the ESP taught over the research period on performance on general English proficiency tests, we also conducted a paired samples t-test to compare the scores in the tests taken and grades obtained in ESP in the research period.Each student has two ESP grades, at the end of the first semester (ESP1) and at the end of the second one, i.e. end of the academic year, ESP2.The scores in all three tests (T1, PF and T2) were computed into grades and paired with grades achieved in ESP.These variables were used both as several values and as average values (ESP average, all tests average).Altogether, six paired-samples tests were run: ESP1 vs. T1; ESP2 vs. T2; ESP average vs.T1, PF, T2 average; ESP average vs.T1, ESP average vs.PF; ESP average vs.T2.The results revealed that there is a statistically significant difference in the scores for all three tests and both grades obtained in ESP (p = 0.00).Since the results for all test pairs indicate a statistically significant mean score difference, they indicate that the students performed significantly better in ESP compared to the grades they would have obtained if English proficiency tests had been graded under the same criteria.The ESP grades average around grade 3 (good), and results on T1, PF and T2 are barely gaining average student grade 2 (sufficient).

Performance by gender, education and employment status
In the next step an analysis of performance with respect to gender was conducted on the total scores on three tests and the two test sections.The total number of students who took the Test 1 and Test 2 was 30, with 15 male students and 15 female, whereas 27 students took the Parallel form, where 14 were male and 13 female.According to scores obtained, it is interesting to note that male participants performed better on all three tests.The analysis of mean differences identified that the greatest difference exists on the Test 1 score with the Mean difference of 3.27 points in favour of male students.Somewhat smaller difference was found on the Test 2 score (Mean difference = +2.26),whereas almost insignificant difference was found on Parallel Form score (Mean difference = +0.81).Results analysed by each test section show that males performed better on the lexical tasks in all three tests (L1, LPF, L2), while females performed slightly better on grammatical tasks of Parallel form and Test 2 (GPF, G2).Since on the majority of tasks, both lexical and grammatical, males performed better, the conclusion is that male students in this group showed higher proficiency in English than females which can be attributed to a number of circumstances beyond the scope of this research.
Observing the performance on the two test sections and total scores on three tests, and the type of secondary school completed, the results obtained indicate that in this sample the students who completed grammar school (N = 11) performed better on all tests and both test sections (Test 1, L1, G1, PF, PFL, PFG, Test 2, L2, G2) than the students who completed vocational school (N = 19).After examining differences with respect to studentsʼ employment status, the results show that the mean values for students who are employed (N = 9) are higher only on G2 and Test 2. On all other variables (Test 1, L1, G1, PF, PFL, PFG, L2) the mean values are higher for the full-time students who are not employed (N = 21).
Due to the very small size of the groups, no further analyses regarding gender, education background or employment were conducted.

Overall scores for all three tests by each task
Further data analysis was conducted in respect of the studentsʼ performance in each of the seven tasks based on the frequencies of task items answered correctly.The average scores for all three tests taken (T1 N = 30; PF N = 27; T2 N = 30) are presented in Table 6.As shown in the table, students performed slightly above average on the Task 5 and Task 6, while on all other tasks they performed below the average result for that task.
Since seven test tasks comprised a different number of items (6 to 10), the percentage of scores for right answers was used to examine and compare overall performance on each particular task.In order to better illustrate the differences among tasks, the average scores are shown in both Figure 5 and Table 7.The percentage of task items done correctly are ranked and shown in Table 7.The highest score was obtained on Task 5 with 55.9 % of the task items answered correctly.Students were asked to choose one of the multiple-choice items on the Present Perfect or Past Simple with four options.The second highest score was obtained on Task 6 with 53.6 % of the correct answers.In this task multiple-choice items with four options were offered on the Present Simple or Present Continuous.First-ranked Task 5 and second-ranked Task 6 are tasks that test grammatical competence.The score obtained on Task 2 with 48.3 % of the correct answers is ranked third and it is the best done task of the Lexical tasks section.The scores obtained on Task 7 in which students were asked to insert a correct relative pronoun occupy the fourth place.It is followed by the scores obtained on Task 3 with 44.1 % of the correct answers.In this task students were offered 10 prepositions to complete the text with these prepositions.The scores of Task 1 are ranked second to last with 36.2 % and Task 4 with 31.6 % is ranked last.The observed results on all three tests indicate that overall the students in this sample performed better in grammatical tasks compared to performance in lexical tasks.
These fi ndings do not support the initial assumption that students would perform better on lexical tasks and the established hypothesis that substantial improvement would be achieved in lexical competence.Test results for students attending ESP classes in the 2012/2013 academic year show an upward trend in favour of their grammatical competence.

Overall scores for each of the three tests by each task
In the next step of data analysis, scores in each respective task were compared across tests (Test 1, PF and Test 2).The score distribution per task is shown in Figure 6.The comparison of scores obtained in each task on all three tests revealed that the students performed better on the Parallel form in six tasks as compared to Test 1 and  The paired samples t-test was conducted to compare scores obtained on each task (Task 1 to Task 7) and scores on the three tests in particular: T1 vs. PF, PF vs. T2 and T1 vs. T2.The paired samples t-test revealed a statistically significant difference between the following items: (Table 8) As can be seen from Table 8, there is a significant difference in the scores for Task 1 of the Test 1 and Parallel form.The comparison of scores for Task 4 revealed a statistically significant difference on two paired samples: Test 1 and Parallel form and Parallel form and Test 2. Likewise, a significant difference was found between scores in Task 5 of the Parallel form and Test 2. In respect of Task 7 significant difference has been found between scores in Task 7 of the Test 1 and Parallel form.On Tasks 1 and 5 significant differences have been found, but these are below one point, therefore not considered of great importance.It is interesting to note that Tasks 1 and 4 belong to test section which examined lexical competence, while Tasks 5 and 7 tested grammatical competence.
Further analysis of mean differences reveals that the greatest difference exists between the performances on Task 7 of the Test 1 and Task 7 of the Parallel form (Mean difference = -1.81 in favour of PF).A somewhat smaller mean difference was found between the performances on Task 4 of the Parallel form and Test 2 (Mean difference = 1.33 in favour of PF) and Task 4 of Test 1 and Parallel form (Mean difference = -1.18 in favour of PF).These results are consistent with the findings on students overall performance which was highest on the Parallel form.

CONCLUSION
The aim of the study was to examine the interrelatedness between learning ESP and general English proficiency.To that end three general English proficiency tests were administered and two major assumptions made: 1) Learning ESP can improve studentsʼ general English proficiency.
2) There is a more substantial improvement in lexical competence as compared to the improvement in grammatical competence.
The analysis of the results reveals that at the overall level the studentsʼ performance over the research period shows an upward trend.Compared to Test 1, more than two thirds (81.5%) of the students scored higher on Parallel form and 60% of the students scored higher on Test 2. With regard to lexical and grammatical competence, a very small, almost insignificant decline in scores has been found on the lexical section and a small increase on the grammatical section of Test 2. These findings do not support the second research assumption that students will substantially improve their lexical competence.Thus, it is reasonable to assume that studentsʼ grammatical competence has been crystallized through studentsʼ participation in the ESP lectures and their use of both grammatical and lexical resources of the English language.The lack of improvement in lexical competence evidenced here may be attributed to studentsʼ focusing on and learning of specific technical terminology used in the vocational field and respective ESP.
A very interesting finding that points to several other perspectives and dimensions regarding foreign language learning is that students performed substantially better on the Parallel form in both lexical and grammatical tasks.As previously commented, in addition to the ESP course, this improvement can be attributed to the time of the test administration, but also to some psychological issues such as the low level of anxiety students experienced or the lower levels of stress they were exposed to due to the timing of the test.Therefore, different factors influencing the improvement achieved should also be investigated more closely.
The results of the research indicate that the students performed better over the research period and that the increase in total scores achieved speaks in favour of the assumptions that ESP can improve studentsʼ general English proficiency.Nevertheless, to establish the extent to which the ESP course is effective in enhancing the students' general English proficiency, further research should be conducted on samples that are larger in numbers and being taught ESP in different vocational fields at diverse colleges.

LIMITATIONS OF THE STUDY
There are a few limitations that might influence the results of the presented research.1.The size of the sample.A group of thirty participants is regarded a minimum group to conduct any quantitative research.Therefore, all findings refer exclusively to this sample and no generalisation can be made thereon.The effect of the studentsʼ gender and other demographic characteristics were not analysed due to the sample size.
2. Administrated tests were not graded.The fact that performance i.e. results on the three administered tests were not graded and had no direct impact on the ESP grades, might have adversely influenced the studentsʼ motivation to do their best.3. Time of the test administration.At the time of taking Test 2, students were exposed to greater stress related to taking and passing exams in other courses.It is reasonable to expect a lack of motivation and reluctance to take an extra test.

Figure 1 :
Figure 1: Representation of Languages Learned (number of students)

Figure 2 :
Figure 2: Foreign Language Learning -Average Duration in Years and Grades

Figure 4 :
Figure 4: Performance in Lexical and Grammatical Tasks on all Three Tests 1 to the lexical section of the Parallel form (t (26) = -3.173,p = 0.004).

Figure 5 :
Figure 5: Percentage of Correct Task Scores Lexical tasks: coloured dark, grammatical tasks: coloured light

Figure 6 :
Figure 6: Distribution of Task Scores on all Three Tests Students performed slightly worse only on Task 5 (Present Perfect vs. Past Simple) on the Parallel form.Comparing task scores on T1 and T2, the students performed better in four tasks (Task 1, 2, 5, and 7) on Test 2.

Table 1 :
An Overview of EGP and ESP Key Characteristics

Table 2 :
Test Tasks by Type

Table 3 :
Descriptive Statistics of the Three Tests

Table 6 :
Average Scores Obtained on all Seven Tasks