A study of the differences among EFL/ESL methods for reading Comprehension and language awareness

The present paper reports on the effectiveness of the Grammar-Translation Method, Communicative Language Teaching and Content and Language Integrated Learning upon reading comprehension and language awareness. The study recruited a convenience sample of 164 Spanish students from a rural area in southern Spain who randomly received these three ESL/EFL methods. The research was focused on two hypotheses. Firstly, L2 reading learning is more effective from CLT and CLIL than GTM. Secondly, CLT and CLIL demand more contextual learning. Results showed CLT method offered more similar results among the reading comprehension tasks (p<.01). However, from a descriptive study, GTM obtained the best performance globally. In all three methods, scores were weakest for pragmatic awareness. However, the interaction between reading comprehension and language awareness suggested that the intervention was more supportive for CLT. Also, significant intra-gender differences according to the learning methodology were found. Study limitations and future research recommendations also were discussed.


11
The drawback felt in this particular analysis is due to the fact CLT and CLIL are nowadays the cutting-edge of foreign language learning that involves a global process of Reading in connection with other language skills. However, more often than not, evaluation of L2 in the Spanish educational system just covers written skills. GTM emphasizes just visual skills such as Reading and Writing by translating. Many teachers, then, choose a GTM learning that promotes a more successful achievement according to the tests of Spanish educational institutions. Therefore, although GTM appears as a still-viable method, most articles about it come all from lower tier journal considering it as an old method.
Indeed, the present research involved in performing a quasi-experimental study about the efficacy of ESL/EFL methods (GTM, CLT and CLIL) over reading comprehension skills for Spanish-speaking English language elementary students in a rural area. Rural areas are usually characterized by a lack of social and cultural opportunities (Fusarelli, Fusarelli & Riddick, 2018). This is because the population attending rural schools usually come from low-income families. For that reason, many young adults migrate away (Sherman & Sage, 2011). In these areas, it is important to support educated parents' perceptions of school achievement and understand what motivates them to improve their children's future. Thus, hopes for the future serve as the instrumental motivation for gaining proficiency (Noels, Clément & Pelletier, 2001).
Thus, the research questions for the study are the following: (1) Which ESL/EFL method provides the best performance in Reading?
(2) Are there significant differences among GTM, CLT and CLIL methods?
(3) What is the relationship between ESL/EFL methods and language awareness? (4) What is the language awareness that obtained a better performance in each ESL/ EFL method? (5) Are there significant age and gender differences?

Foreign Language Learning Methods
GTM is mainly based on visual and written skills. It demands language transferences between L1 and L2 (Xia, 2014). This approach focuses students' efforts on knowing grammar rules of the target language by comparing native and foreign language in translations. Learning is based on a teacher-centred method. This approach prioritizes intralinguistic components of grammatical awareness (Priya & Jayasridevi, 2018). Nowadays, GTM is considered as a controversial method among many linguists (Zhou & Niu, 2015).
Conversely, CLT is based on students' immersion in exclusively communicative L2 situations (Dolle & Willems, 1984). Its main aim is to help students attain global communicative competence (Natsir & Sanjaya, 2014). A Student-centred learning is prompted and interactions are the key of the teaching-learning process (Nunan, 2004). CLIL methodology was promoted by Marsh (1994). This method is based on using L2, not only with linguistic contents, but also with non-linguistic areas (Agustín, 2015;Jaekel, 2018). The interdisciplinary concept is a key dimension in this approach. Both CLT and CLIL take into account intralinguistic and extralinguistic factors of L2 learning and acquisition.

Language Awareness, Reading Comprehension and Foreign Language Learning
Language awareness consists of two cognitive factors: linguistic knowledge analyses and attention control span (Llombart, 2017). The first of them is related to the semantic value of language regarding the adjustments that the subject performs between signifier and signified (Li & Wu, 2015). This is the first step to shape language awareness (Candry, Deconinck & Eyckmans, 2017).
Regarding the second cognitive factor, the attention process is classified into three types: divided, held and selective attention (Stavrinos et al., 2018). It is a basic psychological function that integrates executive functions that manage learners' behaviours (Rebollo & Montiel, 2006). The attention process is produced by a continuum of meaning distributed into different syntactic constructions that determine reading comprehension. This is the first step for the decoding process (Young-Suk, 2015). Hence, language awareness becomes a previous condition for grapheme-phoneme conversion and the recognition of words as a whole to obtain meaning (Li & Wu, 2015). This implies, then, that there is a simultaneous bottom-up and top-down process in reading comprehension (Peterson, 2001).
The balance between decoding and attention in the reading process, on the one hand, drives reading comprehension learning and, on the other hand, counteracts the effects of the overloading strategies transferred from L1 to L2 (Pollard-Durodola & Simmons, 2009). This prompts a common linguistic shaping between L1 and L2 to enable transferences of learning to learn strategies from one language to another (Agustín, 2015). Therefore, the activation of understanding frames in texts (Fernández & Montero, 2015) and interpreting strategies are developed (Amenós-Pons, Ahern & Guijarro-Fuentes, 2017).
On that assumption, as Reading is an exchange of graphemes and lexical components organized into semantic values (Núñez-Vázquez & Crismán-Pérez, 2017;Young-Suk, 2015), there are several correlative factors between reading comprehension and language awareness associated with Reading as a cognitive and visual action (McKoon & Ratcliff, 2018). This can be described by two approaches: intrinsic and an extrinsic one.
The intrinsic approach focuses on language awareness. The first step for reading and writing comprehension is the phonological module, which is linked to an orthographic store (Breadmore & Carroll, 2016) produced by the correspondence of graphemes with sounds as significant units. This is particularly relevant for the development phonological awareness in reading.
Most studies state morphological awareness and its relationship with different linguistic disorders as well as the contribution of this module to global language awareness as an organization of grammatical meaning Critten et al., 2014;Goodwin & Ahn, 2013;Vernice & Pagliarini, 2018). On that assumption, morphological awareness may be more closely related to GTM because of the importance of grammar as a central core of the teaching-learning process (Sapargul & Sartor, 2010).
Finally, we emphasize semantic recognition of word categories through the correspondence of lexical units and their visual identification. This module is related to morphological, lexical and semantic awareness in that process (Breadmore & Carroll, 2016). This approach focuses on prioritizing syntactic frames and semantic categories as well as their relationships with semantic awareness. This has been confirmed (Zipke, 2007;Zipke, Ehri & Smith, 2009) with significant correlations between reading comprehension and lexical areas (r=.77, p<0.01). Yet, the acquisition of vocabulary is essential to gain access to reading comprehension (Llombart-Huesca, 2017) especially in the CLIL approach because vocabulary from non-linguistic areas is included (Marsh, 1994). In contrast, meaning determines the place in the syntactic structure that the word or lexical voices occupy (Aragón et al., 2013).
With regard to an extrinsic approach of language awareness, this describes minor skills as predictors of linguistic knowledge. In that way, meaning deduction is underlined (Harmer, 2010), which is crucial in CLT. This process involves meaningful analyses, attitude and opinion inferences as well as discourse patterns and markers.
Indeed, Mertz and Yovel (2009) add pragmatic awareness to language awareness. This consists of several elements: the integration of the sender's intention, situational factors and contextual knowledge in a communicative process where CLT may further prompt meaning and pragmatic factors in a communicative situation. CLT and CLIL demand more contextual learning, so they take into account pragmatic awareness as predominant language awareness, whereas GTM only underlines phonological and morphological awareness (Jafari & Rad, 2016).

Participants
The study recruited 164 schoolchildren, 53 % were boys and 47% girls, aged 10-14 years of age (average age 12.1, SD = 0.81). All of them had Spanish as L1 and English as L2. Children attended three different schools from the same rural area. They all were in the last stage of Elementary Education (6 th grade). They attended English lessons 3 hours weekly. All students followed the same curriculum of foreign language learning but from three different approaches according to the goals of the study. From these students, 28 received GTM; 87, CLT; and 49 CLIL methods. These ESL/EFL lessons were embedded in the regular school curriculum by three English teachers. The teachers who were in charge of performing the target instruction access the educational system by a State Exam that takes as basic requirement a B1 certificate (Council of Europe, 2001). All of them were trained in Communicative Language Teaching to pass the exam but they had freedom to follow their own school plan in class.
Once the target research was planned, they were randomly given charge of each different ESL/EFL method (GTM, CLT and CLIL). Each instructional program took one academic year. The intervention was undertaken during the 2016-2017 school year (30 weeks, 3 hours weekly) according to the school regular curriculum timing. Each ESL/EFL method instruction lasted for a total of 90 hours.
The study was conducted according to the Declaration of Helsinki. Education Department of the Andalusian Goverment (Spain) passed the present research. Educational authorities provided written informed consent. These documents include sponsors' information (the Department of Education of the Andalusian Government, Spain), institutional affiliation (the University of Cádiz), the scope and purpose of the investigation, and the study's benefits from educational activities. Confidentiality was preserved.

Instruments
The TAELIS (Test for the Assessment of English in School; Santamaría 2008) measures English as a second/foreign language in the context of elementary and secondary education. The internal consistency of the test for the target sample was 0.7 according to Cronbach's coefficient alpha. The instrument is administered at the end of the school year after the completion of the instructional programme. The test presents written and oral skills sections, but only the written skills were examined. This section was organized into seven tasks. The tasks used different format to appraise different reading comprehension skills according to some multiple answers. According to their layout, items were as follows: Task 1: items 1-10: filling gaps in sentences. Task 2: items 11-14: visual and semantic equivalence activities. Task 3: items 15-17: semantic equivalence in sentences. Task 4: items 18-26: filling gaps in a text. Task 5: items 27-31: choosing a question according to a previous answer. Task 6: items 32-36: choosing an answer according to a previous question. Task 7: items 37-41: filling gaps in dialogues.
The main distinction among these tasks was in the discernment of situational and contextual factors, which was used to measure reading comprehension. Finally, all items were classified as one of two categories, grammatical knowledge (twenty-one items) or semantic-situational knowledge (twenty items), according to the pre-eminent place of either grammar structure or meaning.

Data Analysis
Statistical analysis was conducted using IBM SPSS version 22.0. Both a descriptive and a correlational analysis were carried out as well as an inferential analysis. Each ESL/ EFL method was described as a single result and comparing means were tested. A correlational study was also examined to study the homogeneity among reading comprehension tasks and its global performance. The correlational study was based on Pearson's correlation coefficient. Afterwards, comparing means of global reading comprehension among ESL/EFL methods were based on the ANOVA test. A post hoc analysis was run by the Tukey Test to find out which ESL/EFL method was different in a pair-wise comparison. Age and gender differences were also studied.
The sample size was N=164 (28 + 87 + 49). This ensured a sampling error of less than 8%. Moreover, 95% confidence interval was established in all contrasts. This delimited Type 1 or false positive error (0.05).

Descriptive study
Descriptive results of reading comprehension were studied ( Table 1). The global score for reading comprehension tasks conformed to the information provided by the typified population of TAELIS test. The typified mean for 6 th grade students in reading comprehension tasks was 29.4 (SD: 20.9). The target sample accomplished a global score of 17.1 (SD: 4.7), results, therefore were inside the range provided by the typified sample with a more homogeneous result according to the standard deviation value.
From a descriptive point of view, results showed the highest mean for task 2 Visual and semantic equivalence activities in GTM and CLT. The CLIL method (Table 4) achieved the highest mean in task 3 Semantic equivalence in sentences. The lowest score for the three ESL/EFL methods was task 6 Choosing an answer according to a previous question (GTM, 0.35; CLT, 0.35 and CLIL 0.31). Yet, all three also agreed the second lowest score (GTM, 0.40; CLT, 0.40 and CLIL, 0.34) in task 4 Filling gaps in a text.

Correlational Study
Pearson's correlation coefficients are presented in Tables 2, 3 and 4). CLT was distinguished by its significant correlation coefficient under .01 (Table 3). However, GTM and CLIL (Tables 5 and 7, respectively) miss the relationship with task 7 Filling gaps in dialogues.

Comparing Means
The results of the ANOVA test were varied. Previously, assumption for homogeneity of variances for the Levene Test was confirmed [Levene (2, 161) =0.667, p>.05].
The ANOVA test showed no significant differences. However, task 4 Filling gaps in a text shows significant differences among ESL/EFL methods [F (2, 161) =3,671, p=.02]. A post hoc analysis of this task was run by the Tukey Test. Finally, significant differences between CLT and CLIL methods in task 4 (p=.05) were provided.
Comparison means by age levels and gender were statistically non-significant. Some differences were found in an intra-gender study. According to results from male participants, task 4 Filling gaps in a text was just significant [F (2, 84)=3,288, p=.04] but not for the post hoc analysis with the Tukey Test. Previously, assumption for homogeneity of variances for the Levene Test was confirmed [Levene (2, 84) = 0.575, p>.05].
The post hoc analysis found that task 3 provided a significant difference between means scores, on the one hand, CLT and GTM (p=.01) and, on the other hand, CLIL and GTM (p=.03). In a descriptive way, GTM obtained the highest mean, 0.75 (SD. 0.28), followed by CLIL, 0.48 (SD. 0.28) and CLT, 0.44 (SD. 0.32) in task 3.
Regarding task 6, the post hoc analysis was significant in a pair-wise comparison for CLIL and GTM (p=.055). Again, GTM had the best result (x̄=0.53, SD. 0.21).

dISCUSSIoN
Reading comprehension has been analyzed from three different ESL/EFL methods in a rural area. Following Wengelin & Arfé (2018), reading comprehension learning can be described in a dual way: as a separate process or in an integrated way in respect of language awareness. This research involves both of them according to different instructional methods.
We found that descriptive results showed that GTM obtained the best global result in reading comprehension (Research Question 1). However, the three ESL/EFL methods did not show a significant difference between mean scores except for CLT and CLIL regarding Filling gaps in a text (task 4, p=.05), which is the most frequently used guided writing in elementary education (Núñez-Vázquez & Crismán-Pérez, 2018). Thus, CLT outperformed CLIL. Additionally, CLT provided the highest correlational performance among all reading comprehension tasks. This means learning was more homogeneous for the CLT method globally due to the fact that it showed more similar results among the different reading tasks. Therefore, CLT outperformed the other ESL/EFL methods (Research Question 2).
In this research, the three ESL/EFL methods behaved in a similar way regarding language awareness (Research Question 3). Thus, the best results were obtained in tasks of intra-linguistic components. On the one hand, these components are associated with phonological and morphological awareness which were supported by visual and auditory inputs (Wengelin & Arfé, 2018). On the other hand, intra-linguistic components related to semantic awareness (task 6 Choosing an answer according to a previous question) and extra-linguistic components related to pragmatic awareness obtained the lowest performance (task 4 Filling gaps in a text and 7 Filling gaps in dialogues).
Students from both GTM and CLT achieved their best scores in task 2 Visual and semantic equivalence activities, closely followed by CLIL students. Tasks 2, 4 and 7 were classified as extra-linguistic components of language awareness due to their predominance of communicative and situational factors. This result seems counterintuitive but the layout of task 2 may provide some explanation. This task consisted in decoding visual messages which were printed on a printed signal. These semantic and situational communicative factors were not provided in tasks 6 and 7, not even in task 4 where some contextual information was provided in a written way. From this understanding, communicative information provided by printed signals was more direct to visual information, and, definitely, a conventional ban poster constitutes a specific discursive frame (warning and prohibition), which involves a more intuitive process (Harmer, 2010;Merz & Yovel, 2009;Stavrinos et al., 2018). Put more simply, visual information does not need so much additional semantic and situational communicative information to be interpreted because the key of the communicative process stands on the visual input provided by printed signals which help understanding process (Li &Wu, 2015).
Following the results, pragmatic awareness is the main challenge of ESL/EFL reading comprehension for Spanish-speaking English language students in rural areas. This evidence is confirmed by the need to develop communicative strategies supported by extralinguistic communicative components (Gutiérrez, 2018;Vettorel, 2018). Nonetheless, as mentioned above, CLT showed a more homogeneous performance in the correlational study. This means that L2 linguistic immersion influenced global language awareness.
Participants obtained the highest performance in a task of visual cognitive inputs based on commands for the three ESL/EFL methods, so these results imply the importance of phonological awareness due to the organization or signifiers into meaningful units according to a specific communicative frame. Visual cognitive inputs involve a global language awareness shaping of L1 and L2. The result is corroborated by different research studies which state that L1 contributed to L2 psycholinguistic patterns (Agustín, 2015;Aragón et al, 2013). Thus, results indicated phonological awareness as the basis for language awareness for the three ESL/EFL methods (Candry, Deconinck & Eyckmans, 2017;Li&Wu, 2015). This common language awareness also influences lexical awareness (Kopecková, 2018;Yeung, 2018). Nevertheless, although phonological awareness is a reading comprehension predictor, several studies state that it is not sufficient in order to explain reading comprehension process (Ford et al., 2013) (Research Question 4).
Non-significant differences were also provided in an inter-age and inter-gender study (Research Question 5). However, the relevance of pragmatic awareness is also confirmed by the significant comparing means between CLT and CLIL (task 4 Filling gaps in a text) in an intra-group study for males. Nevertheless, female students showed significant differences between, on the one hand, GTM and CLT and, on the other hand, GTM and CLIL. This showed that female students established GTM as the extreme of a continuum against CLT and CLIL. Female students interpret learning methods as a linear model of learning where the main difference is the treatment given to linguistic skills. GTM emphasizes learning of written skills while CLT and CLIL have more impact on oral skills in an integrated way.
Finally, certain limitations must be considered when interpreting the results. On the one hand, the samples showed an uneven distribution. This issue may apparently affect the homogeneity of variance assumption. However, evidences were provided thanks to the Levene Test (p>.05). ANOVA is considered robust to moderate departures from this assumption in the research (Keppel, 1991).
Furthermore, the relevant variables in ESL/EFL reading comprehension can be those factors which emerge from teaching contextual situations such as affectivity (Arnold, 2000) and resilience (Tae-Young&Yoon-Young, 2017) which characterize rural areas. This context is characterized by constant outmigration so it may be important to examine the processes by which rural children come to be seen as future stayers or leavers because this influences their English level of proficiency and the ways in which individual and family attributes interact with community-level social constructions to influence educational trajectories (Sherman & Sage, 2011).
The variable class-size might also be also playing as a moderator in the present research due to CLIL participants were organized into one single learning group while participants from the CLT method were organized into two teaching groups. For future research, it would be interesting to study how class size influences ESL/EFL methods (Fountas & Pinnell, 2018;Harfitt & Tsui, 2015).
Finally, although all students follow the same curriculum of foreign language learning from three different approaches, it would be advisable to determine the individual baseline in order to study comparing means in an intra-study level. Thus, lack of individual differences may considerably impact outcomes.

CoNClUSIoN
The current study has set out to provide a unified account of reading comprehension in L2 from three ESL/EFL methods based on the analysis of language awareness in Spanish-speaking English language students in a rural area. Regarding previous approaches working under the assumption that reading comprehension can be worked on as a separate process (Wengelin & Arfé, 2018), the present research intended to show that CLT comprises a more integrated relationship of different reading comprehension tasks as the correlational study revealed. In this sense, relating better reading comprehension performance to CLT, the present paper has further shown the importance of a more global communication competence in school contexts (Natsir & Sanjaya, 2014).
As has become clear from the study carried out, this analysis has strong explanatory power about the importance of integrating different reading skills in the development of reading comprehension for English as L2. As a final remark, the effects of class size as well as role gender are prerequisites if one wishes to grasp the differences among the three ESL/EFL methods. Nonetheless, whereas this research focuses on the importance of reading comprehension for students of English as L2, it does so at the expense of in-depth analyses of each ESL/EFL method according to other different skills as both separate processes and integrated ones.
An important task for future inquiry on the ESL/EFL methods and language awareness will be to thoroughly examine the possible variation of results within urban contexts, thus testing the importance of pragmatic awareness to reading comprehension.
Finally, this raises several questions for future research. Indeed, the need for carrying out a similar study to assess the functioning of another receptive skill such as Listening and also check whether there are significant differences among ESL/EFL methods. Such research might be further expanded to productive skills such as Speaking, Talking and Writing.