Introduction

Tests are generally considered an influential decision-making tool, particularly in today’s competitive educational world where individuals with different capabilities or potential are assessed on the basis of their achievement and abilities (Zhang et al., 2011). This is especially true about high-stakes tests. Stenlund et al. (2017) contend that such tests may lead to highly prized goals like admission to higher education or employment for test takers. Test takers might need certain test taking skills and strategic awareness to be successful in such tests. Their strategic behavior in significant test taking situations has drawn the attention of researchers in language testing field, whether it be English as a foreign language (EFL) or English as a second language (ESL) (e.g., Barkaoui et al., 2013; Murray & Riazi, 2018; Nikolov, 2006), indicating how critical it is to EFL and ESL researchers in various contexts.

A glance at the literature on test takers’ strategic behaviors reveals certain conspicuous gaps. Test taking strategies (i.e., strategy use in test situations) appear to have received far less consideration compared to learning or study strategies both in English and other fields, in both western and non-western contexts. Even in EFL or ESL contexts, the limited studies examining test takers’ strategic behavior in test situations have primarily focused on the skills of reading and writing (e.g., Cohen & Upton, 2007; Murray & Riazi, 2018; Nikolov, 2006). Speaking is a demanding productive skill, which differs from the other language skills particularly due to its communicative nature requiring the speaker to interact with another person spontaneously under real time constraints (Pawlak, 2018). This quality of speaking might elicit varied strategic behaviors from test takers (particularly from those with different attentional or linguistic potentials or strategic awareness). Despite the significance of speaking, the studies on strategy use in speaking tests (particularly in high stakes tests, such as IELTS and TOEFL) have been limited to a handful in non-Iranian contexts (i.e., Barkaoui et al, 2013; Fernandez, 2018; Huang, 2013; Swain et al, 2009).  

Even more importantly, a glance at the literature reveals a lack of previous research specifically comparing the strategic behavior of high and low achieving test takers during a speaking test, in both Asian and Western contexts. In addition, the limited previous studies on strategy use in speaking test situations have failed to explore the quality of test takers’ strategies or differentiate between their effective and ineffective strategies (See Barkaoui et al, 2013; Huang, 2013). Exploring the quality of the strategies employed by high and low achieving test takers (for example, through differentiating strategies into effective and ineffective ones) may contribute to a more profound understanding of how ineffective strategy use can influence test takers’ speaking performance.

Understanding high and low achievers’ strategic performance during a speaking test can assist language teachers in finding the weaknesses and potentials of each group of learners, meeting their strategic needs and better preparing them for high-stakes tests such as IELTS. Researchers may also benefit from results related to high and low achieving test takers’ strategic behavior profile. Differentiating between effective and ineffective strategies employed by high and low achievers can present more reliable information on the type of strategic behavior (i.e., effective or ineffective strategies) influencing the two groups’ test performance. The present study aimed to examine Iranian high and low achieving test takers’ strategic behavior in three speaking tasks of IELTS through stimulated recalls as well as their use of effective/ineffective strategies in this speaking test.

Literature Review

Speaking Skill Test Taking Strategies

 According to Cohen (2006, 2014), researchers involved in language testing divide test taking strategies into three types: 1) language learner strategies, 2) test management strategies, and 3) test-wise strategies. Cohen (2012) defines language learner strategies as those dealing with the ways in which test takers operationalize the four basic language skills as well as some other relevant skills including the lexical and grammatical aspects of a language, and translation. Such strategies help test takers prompt the required knowledge for dealing with a task (Cohen, 2014). Test management strategies are defined by Cohen (2012, 2014) as those strategies that work as conscious actions or thoughts facilitating the test taker’s meaningful response to test items and tasks. The final category of strategies, test-wiseness, assists test takers in responding to test items or tasks through knowledge of the format or characteristics of the test (Cohen, 2012). Unlike language learner and test management strategies, test-wise strategies are regarded construct-irrelevant.

The prior research on test takers’ strategic behavior during a speaking test has been mainly conducted in the last decade or so (Barkaoui et al., 2013; Fernandez, 2018; Huang, 2013; Swain et al., 2009). The research on speaking test strategy use has been conducted mainly on the integrated tasks of the TOEFL iBT Test (e.g., Barkaoui et al., 2013; Swain et al., 2009), with IELTS speaking tasks receiving inadequate attention. The research on strategy use during speaking is also limited to a handful of studies, primarily conducted in western or East Asian contexts (e.g., Barkaoui et al., 2013; Fernandez, 2018). These few studies have not provided any information on a comparison of high and lower achievers’ strategy use during speaking. They rarely present a separate profile of test takers’ effective and ineffective strategies in a speaking test. Finally, a majority of the studies on speaking strategies (e.g., Barkaoui et al., 2013; Fernandez, 2018) have not employed very comprehensive speaking strategy classifications. Such existing gaps in the previous studies have led to the need for the current study.    

In the literature on speaking strategies, there are different taxonomies. Researchers (Barkaoui et al., 2013; Fernandez, 2018) contend that a majority of these speaking strategy taxonomies in the literature are limited to only a few strategy categories and individual tactics (in this study, individual strategies and tactics are used interchangeably). For instance, as valuable as the taxonomy employed by Fernandez (2018) is, it is limited to only eighteen tactics. A survey of the literature on speaking strategy taxonomies reveals that the taxonomy developed by Huang (2013) might be the most comprehensive in the literature with its six strategy categories and ninety tactics. What is more, a majority of the strategy taxonomies employed in previous speaking test strategy studies are mainly suitable for integrated speaking tasks in the TOEFL iBT Test (Barkaoui et al, 2012; Swain et al, 2009) or limited to only one speaking task of IELTS (Fernandez, 2018).

High and Low Achievers

High achievers are defined as students who have received higher grades in content areas or certain achievement tests (Hong et al., 2006), while low achievers are learners with lower grades (Wen & Johnson, 1997). These two learner groups have been found to be different in many different areas including self-regulated strategy use, learning motivation, and self-efficacy. High achieving students have been shown to be better at self-regulation and the use of strategies to deal with task challenges in comparison to low achieving students (e.g., Bai et al., 2014; DiFrancesca et al., 2016; Hong et al., 2009). With regard to motivation for learning, high achievers appear to be more concerned about and interested in learning (Abu-Hamour & Al-Hmouz, 2013; Rao et al., 2000). The third difference can be found in self-efficacy, with high achievers possessing higher self-efficacy levels when compared to low achievers (Guo & Bai, 2019). Additionally, research has indicated that high and low achieving test takers may have different strategic behavior profiles which are related to high achievers’ higher tendency toward more effective test taking strategies (Dermitzaki et al., 2008) or their higher use of deep-level strategies compared to low achievers (e.g., Hong et al., 2006).

The current study varied from previous speaking strategy use and high and low achiever comparison studies in that it was an exclusive comparison of high and low achieving test takers’ strategy use reports in the neglected EFL context. Another difference can be related to the fact that the few studies comparing high and low achieving learners observationally in language test situations are mainly limited to reading and writing skills (Dermitzaki, et al., 2008; Nikolov, 2006; Bai, 2018). The final difference between the present study and the previous ones might be that the current study attempts to present a clear account of high and low achieving test takers’ competence in using strategic behaviors in test situations, by differentiating their employed strategies into effective and ineffective ones. By considering the reported differences between high and low achievers and particularly high achievers’ superiority, in this study the researchers assume that high and low achieving test takers might resort to different strategic behaviors in a productively demanding speaking test situation which may require test takers use all their available resources to spontaneously communicate their responses. The following research questions were formulated based on the objectives of the study: 

1. What strategic behaviors (i.e., strategy categories and tactics) do Iranian high and low achieving IELTS test takers report in the three speaking tasks of IELTS?
2. How do Iranian high and low-achieving IELTS test takers report effective and ineffective tactics in the three speaking tasks of IELTS?

Method

In this study, introspection was used to explore Iranian high and low achieving IELTS test takers’ strategy use. Introspective methods of data collection can assist researchers in gaining access to the mental processes involved in language processing and production, which is impossible with non-introspective methods (e.g., Dörnyei, 2007; Gass & Mackey, 2016).

Participants

The participants of the study were 49 Iranian university students who had recently taken the IELTS Test and whose IELTS scores in the speaking section were between 4 and 8.5. They participated in the research voluntarily and their consent was obtained by the researchers prior to the research (Guo & Bai, 2019). In the current study, the researchers drew on expert test-raters’ opinions to define high and low achievers. They operationally defined the high achieving test takers as those who had scored between 7 and 8.5 on the speaking section of IELTS. On the other hand, test takers who had scored between 4 and 5.5 were classified as low achieving test takers. As a result of the selection criteria, the researchers selected 30 out of the initial 49 participants as high and low achievers. The 30 male and female test takers (with an almost equal gender distribution in the two groups) ranging in age from 24 to 30 (mean age=26) were studying in the Humanities Department.

Instruments

Verbal Report (stimulated recall)

In this study, participants verbally reported their self-observation of speaking strategies that they used in the IELTS speaking session. There are two types of self-observations, stimulated recall and retrospective questionnaires/journal entries/interviews (Gass & Mackey, 2000, 2016). In this research, we used stimulated recall (see also immediate retrospection in Yi’an, 1998), which has been reported to be more reliable than delayed retrospection (Ericsson & Simon, 1987). The higher reliability of stimulated recall may be due to the shorter time lag between task performance and verbalization in stimulated recalls, as compared to delayed retrospection. That is, the time interval is shorter in stimulated recalls, and this may help test takers recall their thinking processes more easily and accurately, resulting in a more reliable verbalization of those thoughts.

Stimulated recalls were also preferred to self-revelation or think-aloud (another common tool in language testing) because of their lower susceptibility to reactivity (Ericsson, 2002). In think-aloud, test takers may experience different internal processes depending on whether they perform a language task with or without thought process verbalization. This may result from dual processing—carrying out a learning task while talking about thinking processes at the same time (Ellis, 2001). However, in stimulated recalls, test takers do not have to perform a language task and talk about the thinking processes simultaneously, which lowers reactivity. The spontaneous nature of the speaking skill, which makes concurrent verbalization/think-aloud impossible (Gass & Mackey, 2016), is another reason for the selection of stimulated recalls. Another rationale is that despite the support for self-revelation or think-aloud (Cohen, 2000), the need for researchers’ creation of the real exam circumstances (Gu & Shi, 2012) may not allow for the concurrent verbalization in the case of studies on test-taking strategy use.

Finally, stimulated recalls need either aural or visual support, with some researchers emphasizing the superiority of visual over aural support (e.g., Dörnyei, 2007; Gass & Mackey, 2016). However, in this study the researchers used audio recording, rather than video recording, during the speaking session for the immediate recall to respect the participants’ preference.   

Three IELTS Speaking Tasks

Three IELTS speaking tasks were used for the study. These tasks are in the form of interviews conducted by an IELTS examiner, and they normally last between eleven and fourteen minutes. In the first task, there is generally an introduction, followed by some questions on general topics. The task typically lasts about four or five minutes. In the second task, the candidate is given a cue card with some questions on a certain topic. The candidate is given one-minute preparation and is then expected to have a monologue on the cue card questions for one to two minutes. In the last task in the speaking section of IELTS, which is a discussion and lasts about four to five minutes, candidates are asked to talk about abstract matters. The speaking task topics were selected randomly from a Cambridge IELTS book, following Fernandez (2018). The topic of the first task was television, and the topic for the second and third tasks was friendship.

Data Collection Procedure

To collect the data, the researchers first invited the participants to take part in the study, and obtained their oral consent. Next, two different versions of the research were conducted. The test takers initially went through a familiarization phase (See Barkaoui et al, 2013; Swain et al, 2009) which helped them become familiar with the process of verbalizing thoughts (i.e., stimulated recalls). This version of the test provided the test takers with an opportunity to practice carrying out stimulated recall after each speaking task before the main phase of research. This phase helped the participants learn how to report their speaking strategies by using stimulated recalls. During the familiarization phase, all the test takers were given the same required instructions on how to do the stimulated recall after speaking tasks based on the literature. After one week the researchers began the research phase (i.e., main phase) of the IELTS speaking task. In carrying out the research phase, the researchers went to considerable effort to make the speaking session similar to the real exam circumstances (Gu & Shi, 2012).

During the research phase, each test taker was asked to participate in the speaking session while their speaking time was audio-recorded. Since the time interval between the speaking session and the stimulated recall is important, the verbalizations were carried out immediately after each of the three IELTS speaking tasks. An audio recording of each test taker’s speaking was played back to them as a stimulus to help them retrieve and verbalize their strategies during the speaking task. They were similarly given the permission to start or to pause replays of the audio-recordings and verbalize the thoughts they had before, during and immediately after each task. The researchers used non-leading questions to prompt them to verbalize their thoughts at particular moments or explain their verbal reports more. Unlike the speaking tasks, no specific time limit was set for the verbalization (Gu & Shi, 2012). The verbal reports were conducted in Persian, the official language in Iran and one which worked like an L1 (Dörnyei, 2007; Gass & Mackey, 2000) for the participants.

Since one of the objectives of this study was to identify effective and ineffective tactics and find their frequency, the researchers developed a 3-point checklist (ineffective, neutral, and effective) to differentiate between and determine effective and ineffective tactics. They drew on Huang’s (2013) 90 tactics (Appendix) along with the definitions and examples of each tactic in order to design the checklist. Next, they asked thirty experienced IELTS teachers to identify the effectiveness of each tactic in the checklist.

Data Coding and Analysis

The researchers audio-recorded the test takers’ speaking sessions and stimulated recalls during the verbal reports for the purpose of data analysis. The spoken data was then transcribed, coded and analyzed to identify the existing strategies. The researchers drew on Huang’s (2013) taxonomy of the six strategy categories of approach, communicative, cognitive, meta-cognitive, affective and social and 90 tactics employed in IELTS speaking tasks (Appendix) to code and analyze the data. It should be noted that based on the preliminary analysis of their data, the authors of the present study divided some of Huang’s (2013) tactics into subcategories.

Coding the data involved certain stages. For the ease of coding the researchers first assigned numbers and codes to Huang’s (2013) taxonomy of speaking strategies, which comprised six categories and 90 tactics. In the next stage, a trained research assistant (RA) coded 33% of the high and low achievers’ verbal report data, using these numbers and codes. Next, the first researcher randomly coded 20% of the same data, using the same coding scheme.

To estimate inter-coder reliability, the researchers followed previous researchers (Fernandez, 2018; Goh, 2002; Huang, 2013; Swain et al., 2009) by using inter-coder agreement percentage. The inter-coder agreement percentage between the RA and the first researcher was 91%. The RA and the researcher discussed and resolved the differences in their coding. Next, the RA continued coding the remaining transcripts. For speaking performance scores, the researchers asked two professional IELTS examiners to independently rate the test takers’ speaking based on the scoring rubrics for IELTS speaking section. The raters discussed their disagreements in scoring until 100% scoring consistency was established (See Swain et al., 2009). Finally, the researchers used frequency counts and percentages to answer both the first and the second research questions.

Results

In the following, the results concerning the Iranian high and low achieving IELTS test takers’ overall use of strategies (i.e., strategy categories) and tactics (i.e., individual strategies) are presented.   

Research Question #1: What strategic behaviors (i.e., strategy categories and tactics) do Iranian high and low achieving IELTS test takers report in the three speaking tasks of IELTS?

As Table 1 indicates, high and low achieving test takers in this study reported a total of 1913 tactics and six strategy categories in their verbal reports. Out of this total, high achievers reported 1075 (56.2%) tactics whereas low achievers reported only 838 (43.8%).

Table 1: Low and high achievers’ strategic behaviors across strategy categories

The high and low achieving test takers had both similarities and variations in terms of the hierarchy of strategy category use, and rather considerable variations in the frequency of these categories. For low achievers, meta-cognitive (12.9%) and communication (10.6%) strategies had the largest percent of those reported, followed by cognitive (9.09%) and approach (5.07%) categories. The least frequent strategy categories in the low achievers’ reports were affective (3.39%) and social (2.66%) strategies. Similarly, the two most frequent strategy categories in the high achievers’ verbal reports were meta-cognitive (17.3%) and communication (14.6%) strategies, followed by approach (9.36%) and cognitive (9.04%) categories. Affective (4.23%) and social (1.67%) strategies were the least frequent in the high achievers’ verbal reports, similar to the low achievers’ data. Table 1 also indicated some variations in the frequency of strategy categories between the groups, with high achievers reporting more tactics in each category.

In addition to analyzing categories of strategy use, the researchers analyzed high and low achievers’ use of individual tactics. A detailed analysis of the occurrences of the various tactics within Huang’s (2013) six strategy categories is presented in Table 2.

Table 2: High and low achievers’ tactics use

Overall the high and low achieving test takers employed 84 out of the 90 tactics available in Huang’s (2013) taxonomy, leaving six tactics to be missing in the data, as indicated in Table 2. The missing tactics were from different categories.

Table 2 also indicates both similarities and variations between the two study groups in terms of tactic use hierarchy/order, while in terms of tactic use frequency the variations are more outstanding than resemblances. For instance, the tactics most commonly reported by the two groups in the approach category (generating choices, making choices and generating ideas) were reported in a similar order by both high and low achieving test takers. However, the two groups differed in the frequency of the tactics, with the high achievers reporting more of such tactics compared to low achievers.

Research Question #2: How do Iranian high and low-achieving IELTS test takers report effective and ineffective tactics in the three speaking tasks of IELTS?

The second research question aimed at the test takers’ effective and ineffective tactics. The usefulness or appropriateness of the strategies was determined based on thirty expert IELTS teachers’ answers to a 3-point checklist. Table 3 indicates the high and low achievers’ use of effective and ineffective tactics, as reported during their stimulated recalls.

Table 3: Effective and ineffective tactics reported by high and low achievers    

As Table 3 indicates, 1,198 (62.6%) out of the total of 1,913 tactics reported by the high and low achieving test takers were defined either as effective or ineffective, while 37.4% were defined as neutral. The high and low achieving test takers appeared to report different levels of effective strategies. The high achievers reported 558 (46.6% of the total tactics used) whereas the low achievers reported only 308 (25.7% of the total tactics used). On the other hand, the two test taking groups indicated minimal variations in the use of ineffective tactics. That is, 185 ineffective tactics (15.4% of the total tactics used) were reported by the low achievers, with the high achievers reporting 147 (12.3% of the total tactics used).

Discussion

The current study was conducted with thirty high and low achieving IELTS test takers in the Iranian context to discover their strategic behaviors (i.e., self-reported use of strategy categories ), and their use of particularly effective and ineffective tactics. The results indicated that the high achieving test takers reported more strategies and tactics during simulated recalls than their low achieving counterparts. This finding may first corroborate the literature implying high achievers’ heavier reliance on strategic behaviors compared to low achievers (Dermitzaki et al., 2008; DiFrancesca et al., 2016). It seems that high achieving test takers are more strategic in their test situation behavior than their low achieving counterparts. The finding may indicate high achievers’ higher levels of strategic awareness during tests. When high achievers engage in a testing situation, they might be mentally better prepared for the challenges of a demanding test compared to low achievers, and able to consciously use their available linguistic and attentional resources to deal with the challenges. It might also be the case that in practice, high achievers understand the significance of strategic competence as part of their communicative competence for the successful completion of a communicative task.

The results revealed some similarities between high and low achievers in the hierarchy of strategy category reports. Evidence of these findings can be seen in previous studies indicating metacognitive (e.g., Barkaoui et al, 2013; Huang, 2013) and communication (e.g., Huang, 2013; Swain et al, 2009) strategies as highly reported strategy categories during the speaking section of high-stakes tests such as IELTS and TOEFL.

The findings also lend support to other studies indicating the lower frequency of affective (Swain et al, 2009) and social (Huang, 2013) strategies in the speaking section of high stakes tests such as IELTS and TOEFL. This finding might imply the existence of a hierarchy of strategy significance. It seems that certain strategy categories (such as the meta-cognitive and communication categories in this study) are essential for test takers in test situations, irrespective of test takers’ achievement levels. On the contrary, other strategies (such as the social and affective categories in this study) might serve test takers, whether high or low achiever, less. This may possibly have to do with the educational environment, or more specifically (the IELTS) classroom structure in the Iranian context, where certain strategic behaviors might be nurtured in (EFL) learners more than others, whether directly or indirectly. However, this assumption may require in-depth observational studies of IELTS teachers’ classrooms.  

The results indicated rather considerable variations between high and low achievers in the frequency of meta-cognitive, communication, approach and social categories. As stated earlier, metacognitive and communication strategies might serve as a must in test taking situations. There might be two reasons for the higher frequency of these categories in the high achievers’ verbal reports. Since these strategy categories are important (e.g., Barkaoui et al, 2013; Huang, 2013), it is reasonable to expect high achievers who are strategically aware and competent test takers to use them more. The speaking section of IELTS is a complex test taking situation and it may be logical to expect the more competent test takers like high achievers to use more strategies than the less strategically conscious and less well-prepared test takers.

In high-stakes testing situations, more successful test takers may employ strategy types capable of assisting them in dealing with complex tasks. It seems that the Iranian high achieving test takers were more conscious of orienting themselves to the task, which could be a reason for their heavier reliance on approach strategies compared to low achievers. This might have been particularly due to high achievers’ heavier reliance on the decision related tactics of generating and making choices, which in turn might attest to high achievers’ possible preference for better choices during tests. Interestingly, one strategy category which low achievers reported using more frequently than high achievers was the social category. This may be rooted in low achievers’ undue concern with the examiner while performing the speaking test, and most probably due to their frequent resort to seeking clarification from the examiner. 

Tactic use analysis indicated similarities in the hierarchy of most frequently reported tactics by high and low achievers and variations in the frequencies of such tactics. This might mean that certain tactics may serve as a must for the speaking section of the IELTS test, irrespective of high and low achievement. For instance, seeking clarification can be considered a common tactic for both high and low achievers in the social category. However, the higher frequency of tactics in high achievers’ data seems to be rooted in high achievers’ higher level of strategic competence and awareness, which possibly results in their heavier reliance on essential and common tactics compared to low achievers. Additionally, our results indicated high achievers’ higher use of effective strategies in comparison to low achievers. In contrast, low achievers used ineffective strategies more than high achievers. This finding matches those of other researchers (Dermitzaki et al., 2008; DiFrancesca et al, 2016) reporting high achievers’ greater use of effective strategies in test situations and low achievers’ reliance on mainly ineffective strategies. Compared to other findings in this study, this finding may help explain high achievers’ higher levels of strategic competence or awareness, and by contrast, low achievers’ possibly incomplete strategic competence. High achievers seem to better understand what tactics and strategies to use when faced with a challenging test experience, such as the speaking test with its uniquely spontaneous nature, and this knowledge may result in their greater use of effective strategies to successfully deal with test challenges. Additionally, high achievers’ greater use of effective strategies as well as low achievers’ greater use of ineffective strategies might have possibly been another reason why high achievers outperformed low achievers.     

Conclusion

In the current study there were certain limitations that may reduce the generalizability of the findings. These limitations may include the impossibility of the verbalization of certain strategies by nature even despite the use of stimulated recalls, test takers’ selectivity in strategy report, the exclusion of average achievers in the study, as well as a lack of correlational design to understand the link between strategy usage and test scores. Future studies in both Iranian and non-Iranian contexts can provide a more accurate understanding of high and low achieving test takers’ strategic behavior profiles by considering these limitations.

Despite the limitations, certain conclusions and implications for students, teachers and major decision makers in the field of language education can be offered in the present study. The findings imply potential variations in high and low achieving test takers’ strategic behavior profiles. High achievers might be strategically more conscious of the processes they go through during a test and may employ their available strategies more than low achievers. Having understood the difficulties of the test situation, high achievers may look for more effective strategies to deal with the challenges while their low achieving counterparts may resort to ineffective strategies, resulting possibly from their incomplete strategic competence. Further, certain strategy categories and individual learner strategies (tactics) might be similarly used by high and low achieving test takers. This is because such strategies are more essential for a test situation like a speaking test. 

The findings may have implications for students, teachers and major decision makers such as curriculum and syllabus designers. Since there were variations between high and low achievers in their strategic behaviors (i.e., high achievers’ strategically more successful behavior), teachers may want to consider training and raising awareness of strategies in their classrooms. This is particularly suggested in high-stakes test-preparation classrooms, and for low-achieving students; low achieving students should work on improving their strategic behavior and enhancing their strategy knowledge. In the study low achievers indicated a heavy reliance on ineffective strategies, which means that strategy use awareness should be raised in such students. Finally, major decision makers such as curriculum and syllabus designers may want to focus on strategic competence as a significant part of communicative competence and even possibly include effective strategy application in the scoring system of high-stakes tests like IELTS or English syllabus or curriculum.

Acknowledgements

The authors would like to thank Dr. Li-Shih Huang, the participants of the research, and the anonymous reviewers for their valuable help. 

References

Abu-Hamour, B., & Al-Hmouz, H. (2013). A study of gifted high, moderate, and low achievers in their personal characteristics and attitudes toward school and teachers. International Journal of Special Education, 28(3), 5-15.

Bai, B. (2018). Understanding primary school students’ use of self-regulated writing strategies through think-aloud protocols. System 78, 15-26. https://doi.org/10.1016/j.system.2018.07.003

Bai, R., Hu, G., & Gu, P. Y. (2014). The relationship between writing strategies and English proficiency in Singapore primary schools. The Asia-Pacific Education Researcher, 23(3), 355-365. https://doi.org/10.1007/s40299-013-0110-0

Barkaoui, K., Brooks, L., Swain, M., & Lapkin, S. (2013). Test-takers’ strategic behaviors in independent and integrated speaking tasks. Applied Linguistics, 34(3), 304-324. https://doi.org/10.1093/applin/ams046

Cohen, A. D. (2000). Exploring strategies in test taking: Fine-tuning verbal reports from respondents. In G. Ekbatani & H. Pierson (Eds.), Learner-directed assessment in ESL (pp. 131-145). Erlbaum.  

Cohen, A. D. (2006). The coming of age of research on test-taking strategies. Language Assessment Quarterly, 3(4), 307-331.https://doi.org/10.1080/15434300701333129

Cohen, A. D. (2012). Test-taking strategies. In C. Coombe, P. Davidson, B. O'Sullivan, & S. Stoynoff (Eds.), The Cambridge guide to second language assessment (pp. 96-104). Cambridge University Press.

Cohen, A. D. (2014). Using test-wiseness strategy research in task development. In A. J. Kunnan (Ed.), The companion to language assessment (pp. 893-905). Wiley/Blackwell.

Cohen, A. D., & Upton, T. A. (2007). ‘I want to go back to the text’: Response strategies on the reading subtest of the new TOEFL. Language Testing, 24(2), 209-250. https://doi.org/10.1177/0265532207076364

Dermitzaki, I., Andreou, G., & Paraskeva, V. (2008). High and low reading comprehension achievers’ strategic behaviors and their relation to performance in a reading comprehension situation. Reading Psychology, 29(6), 471-492. https://doi.org/10.1080/02702710802168519v  

DiFrancesca, D., Nietfeld, J. L., & Cao, L. (2016). A comparison of high and low achieving students on self-regulated learning variables. Learning and Individual Differences, 45, 228-236. https://doi.org/10.1016/j.lindif.2015.11.010

Dörnyei, Z. (2007). Research methods in applied linguistics. Oxford University Press.

Ellis, R. (2001). Introduction: Investigating form-focused instruction. Language Learning, 51(1), 1-46. https://doi.org/10.1111/j.1467-1770.2001.tb00013.x

Ericsson, K. A. (2002). Toward a procedure for eliciting verbal expression of nonverbal experience without reactivity: Interpreting the verbal overshadowing effect within the theoretical framework for protocol analysis. Applied Cognitive Psychology, 16(8), 981-987. https://doi.org/10.1002/acp.925

Ericsson, K. A., & Simon, H. A. (1987). Verbal reports on thinking. In C. Færch & G. Kasper (Eds.), Introspection in second language research (pp. 24-53). Multilingual Matters.

Fernandez, C. J. (2018). Behind a spoken performance: Test takers’ strategic reactions in a simulated Part 3 of the IELTS speaking test. Language Testing in Asia, 8(18), 1-20. https://doi.org/10.1186/s40468-018-0073-4

Gass, S. M., & Mackey, A. (2000). Stimulated recall methodology in second language research. Erlbaum.

Gass, S. & Mackey, A. (2016). Stimulated recall methodology in applied linguistics and L2 research. Routledge.

Goh, C. C. M. (2002). Exploring listening comprehension tactics and their interaction patterns. System, 30(2), 185-206. https://doi.org/10.1016/S0346-251X(02)00004-0

Gu, X., & Shi, C. (2012). A retrospective study on test-takers' cognitive and metacognitive processes in taking a compound dictation test. Chinese Journal of Applied Linguistics, 35(4), 400-420. https://doi.org/10.1515/cjal-2012-0030

Guo, W. & Bai, B. (2019). Effects of self-regulated learning strategy use on motivation in EFL writing: A comparison between high and low achievers in Hong Kong primary schools. Applied Linguistics Review. https://doi.org/10.1515/applirev-2018-0085

Hong, E., Peng, Y., & Rowell, L. L. (2009). Homework self-regulation: Grade, gender, and achievement-level differences. Learning and Individual Differences, 19(2), 269-276. https://doi.org/10.1016/j.lindif.2008.11.009

Hong, E., Sas, M., & Sas, J. C. (2006). Test-taking strategies of high and low mathematics achievers. The Journal of Educational Research, 99(3), 144-155. https://doi.org/10.3200/JOER.99.3.144-155

Huang, L.-S. (2013). Cognitive processes involved in performing the IELTS speaking test: Respondents’ strategic behaviors in simulated testing and non-testing contexts. IELTS Research Reports Online Series, 1, 1-51. https://www.ielts.org/-/media/research-reports/ielts_online_rr_2013-1.ashx

Murray, J. C. & Riazi, A. M. (2018). Composing strategies reported by high and low achievers in the TOEFL-iBT integrated task: Implications for investigative pedagogy. Writing & Pedagogy, 10(1-2), 129-160. https://doi.org/10.1558/wap.30570

Nikolov, M. (2006). Test-taking strategies of 12- and 13-year-old Hungarian learners of EFL: Why whales have migraines. Language Learning 56(1), 1-51. https://doi.org/10.1111/j.0023-8333.2006.00341.x

Pawlak, M. (2018). Investigating the use of speaking strategies in the performance of two communicative tasks: The importance of communicative goal. Studies in Second Language Learning and Teaching, 8(2), 269-291. https://doi.org/10.14746/ssllt.2018.8.2.5

Rao, N., Moely, B. E., & Sachs, J. (2000). Motivational beliefs, study strategies, and mathematics attainment in high- and low-achieving Chinese secondary school students. Contemporary Educational Psychology, 25(3), 287-316. https://doi.org/10.1006/ceps.1999.1003

Stenlund, T., Eklöf, H. & Lyrén, P.-E. (2017). Group differences in test-taking behavior: An example from a high-stakes testing program. Assessment in Education: Principles, Policy & Practice, 24(1), 4-20. https://doi.org/10.1080/0969594X.2016.1142935

Swain, M., Huang, L. S., Barkaoui, K., Brooks, L., & Lapkin, S. (2009). The speaking section of the TOEFL iBT (SSTiBT ): Test-takers’ strategic behaviors. (TOEFL iBT Research Report, 10). Educational Testing Service.

Wen, Q., & Johnson, R. K. (1997). L2 learner variables and English achievement: A study of tertiary-level English majors in China. Applied Linguistics, 18(1), 27-48. https://doi.org/10.1093/applin/18.1.27

Yi’an, W. (1998). What do tests of listening comprehension test? A retrospection study of EFL test-takers performing a multiple-choice task. Language Testing, 15(1), 21-44. https://doi.org/10.1177%2F026553229801500102

Zhang, W., Liu, M., Zhao, S. & Xie, Q. (2011). English test-taking strategy use and students’ test performance. Asian EFL Journal, 13(2), 133-168. http://asian-efl-journal.com/PDF/Volume-13-Issue-2-Zhang.pdf