Strategy Intervention and Reading Proficiency
in the Modern Tertiary EFL Classroom

                                                                                                             Stephen English

Abstract
There are critical issues pertaining to current and future challenges for university students and their educators, regarding L2 reading skills, classroom instruction and test taking. At first, this paper explores research into the processes involved for learners of English when accessing L2 texts, and then describes current research directions into learning strategies and reading intervention including web-based literacy instruction. The vulnerability of many students` own lack of awareness to successfully apply reading strategies efficiently and timely, seems particularly relevant to Japanese students, coping with daily reading comprehension tasks both within and outside the classroom setting and who regularly take placement tests.

Introduction
In the current climate of increasing competition between university graduates in Japan, coping with the educational needs of these students is, and will continue to be a challenge for tertiary educators. Particularly with regard to foreign language faculties, these pedagogues’ duties are multi-functional, providing instruction that focuses on all language skills, and that might place emphasis on conversational, grammatical or literacy content. As both entry and exit criteria to and from most universities typically rely upon placement tests based on textual evaluation or multiple-choice TOEIC/TOEFL exams with minimal oral content, high-level literacy skills in particular, continue to be important attributes for Japanese students entering university or the work-place. However, McVeigh`s (2002:100) contention that Japanese students throughout their education, …receive heavy doses of ‘closed-knowledge’ training, characterised by “filling in the blanks”, “memorising” and “separating facts” for example, …and are socialised by the meta-curriculum to associate this style with education, appears in contradiction to …“open-knowledge” literacy features such as connecting information, abstract and critical thinking necessary for higher order comprehension.
This study therefore hopes to identify potential weaknesses in current assessments of all learners` literacy needs, first by presenting a brief historical account of issues for both L1 and L2 readers, leading to factors that could ultimately be important for improving second language literacy skills.

Defining Reading Ability

Reading ability can be defined as the efficiency …to draw meaning from the printed page and interpret this information appropriately (Grabe and Stoller, 2002, p.09). However this definition does not adequately explain affecting variables such as the purposes for reading that will require different skills and strategies, or the criteria involved that explain general reading comprehension skills.  It also does not indicate the varying linguistic and cognitive factors involved such as the metaphorical models of bottom-up, top-down and interactive processes.  Finally with regard to the second language (L2) reader, it fails to explain whether texts are accessed in the same manner as for first language (L1) readers, and show whether meaning accessed is proportional or not to the language proficiency of the L2 reader.

Bottom-up v. top-down processing models
Historically researchers and educators have long regarded reading as a ‘bottom-up’ process (Ellis and Sinclair, 1989), where fluent comprehension entails sequencing language input in an expedite manner. This idea views linguistic processes more as a skill, proportional to the amount of exposure to texts and the efficiency of processing. The more fluent and automatic these processes are, the more skilful and proficient the reader becomes, and within this paradigm, fluent L2 readers are required to have a minimum vocabulary store in long-term memory.
Whereas bottom-up models rely on lower-level linguistic processing, ‘top-down’ models are based on cognitive higher-level processes where emphasis is put on the reader extracting and directing enough information from a text in order to confirm or reject various expectations or prior knowledge (Bartlett, 1932). Here, reading … is a selective process.  It involves partial use of available minimal language cues selected from perceptual input on the basis of the reader’s expectation (Goodman, 1970, p.260).

Inferences and accessing background information are key mechanisms in this process. Inferences are problem-solving devises that help relay plausibility and logicality of incoming text and which Mackay and Mountford (1979) consider important for the preparation and use of L2 reading materials. Important skills include being able to link propositions together as well as identify ambiguous statements.  Examples of logical inferences include making schematic links (Nix, 1983), where information in a first proposition is needed to interpret the second. When combined, these mechanisms help the reader/test-taker to gain access to all levels of textual comprehension ranging from literal comprehension through to more advanced interpretive and critical comprehension. Literal comprehension involves understanding surface meanings, where readers are asked to find information and ideas explicitly stated in the text. Interpretive comprehension involves searching beyond surface meaning. The reader must be able to identify relationships among ideas, drawing conclusions and predicting outcomes. Finally critical comprehension requires students to be able to identify deeper meaning units, for example to differentiate between opinions and facts and assess the accuracy of textual information.

For many years researchers have investigated whether access to meaning emphasises more importance on ‘top-down’ or ‘bottom-up’ processing. Describing reading as a psycholinguistic process, Goodman`s (1973, 1983) ‘Psycholinguistic Guessing Game’ model, views reading comprehension as a repetitious process of hypothesising, sampling and confirming background knowledge.  A good reader is selective with the information available only choosing, …enough to select and predict a language structure, …which is decodeable  (Goodman, 1973, p.164). Goodman’s model is typical of higher-level processes, however it does not adequately explain how a good reader is able to precisely select that information which is useful and that which is not.

Factors affecting L2 literacy

Theoretical approaches to L2 literacy have been divided, with some (Cumming, 1990), taking the position that literacy in a second language is a cognitive function, whilst others (Walace, 1986; Gillespie 1993), see it as a social function.  Some researchers even consider there to be multiple literacy as a function of varying social and discourse contexts (Johns 1997).
There have also been various theories put forward to explain the relationships affecting L2 literacy. Alderson (1984) described two factors that might cause difficulties in L2 reading ability, namely L1 reading and L2 linguistic proficiency, which led towards two opposing hypotheses being put forward. The ‘linguistic threshold hypothesis’ states that a certain threshold of L2 linguistic ability is necessary before L1 reading ability can be transferred to a second language, whereas the ‘linguistic interdependence hypothesis’ allows for any L1 reading ability to be transferred from L1 to L2 regardless of L2 linguistic proficiency.

Product v. process approaches to L2 literacy
With regard to L2 reading comprehension ability, research design has also typically been divided, following one of two approaches. First is to consider reading ability to be a ‘product’ of the number of meaning representations gained over the course of reading texts. On the other hand it might be regarded as being proportional to the types, or ‘processes’, of mental activities whilst engaged in a text in order to construct meaning. This ‘product’ verses ‘process’ oriented approach has been studied in detail and will be considered as follows.

Product-oriented studies are by nature quantitative in design. Tests are given that might shed light on the influence of variables including L1 reading ability, L2 reading ability and L2 linguistic proficiency. Often through statistical multiple-regression analysis, results are then interpreted. Most product-oriented studies have shown a decrease in the relationship between L1 and L2 reading when a reader’s L2 proficiency decreases. Studies including Brisbois (1995) and Lee and Shallert (1997) indicate strong evidence in favour of the linguistic threshold hypothesis, where L1 reading ability only transferred to L2 for high-level readers, and that in the case of low-level readers, transfer was not identified at all. Variations have been significant depending on research design, but large numbers of participants have indicated that the relationship between L1 and L2 reading is weak. Whereas product-oriented studies can gather data over a wide range of participants in one go, qualitative process-oriented studies by design, tend to sample individuals or small groups. This is due, in part, to time constraints imposed by the methodology of the research, where for example, case studies use think-aloud protocols to gain insights into the various individual mental interactions between reader and text. These protocols rely upon a reader to verbalise their mental processes in real time as they proceed through a particular text, and through the researcher’s interpretation of participants thoughts, conclusions are formed. The collection and analysis of such data takes far more time than that required for quantitative product-oriented reading studies.

Particularly interesting to researchers, regarding think-aloud protocols, has and will continue to be those strategies used by readers to gain meaningful interpretations of the texts that they are reading. From understanding process related variables that determine meaning, research has consistently supported the ‘linguistic interdependence hypothesis’, indicating a correlation between L1 and L2 reading ability, regardless of L2 proficiency level. Davis and Bistodeau (1993) provide evidence that L1 reading strategies transfer to L2 for low level L2 readers as well as for high level L2 readers　　and Zwaan and Brown (1996) have shown that L2 readers with high L1 reading ability are more accurate at paraphrasing than those with low L1 ability.

Accessing Texts Through Learning Strategies

Reading strategies mentioned above, are part of overall ‘learning’ strategies, consisting of learning behaviours, problem solving and study skills that can facilitate more efficient and effective learning, and in the case of reading, help facilitate comprehension. Chamot (1987, p.71) describes learning strategies as,…techniques, approaches or deliberate actions that students take in order to facilitate the learning and recall of both linguistic and content area information. The distinction between learning and reading is an important one to consider. An L2 learner accessing a text might activate learning strategies (for example, cognitive strategies to recall words expeditiously), as well as comprehension (reading) strategies, that help gain meaning of textual information. Cognitive strategies are one of four categories of learner strategies identified by O’Malley and Chamot (1990), alongside metacognitive, social, and affective strategies.  Oxford (1990) further sub-divides cognitive strategies, to include ‘memory’ and ‘compensation’ strategies, and as shown below, includes them as part of a detailed taxonomy of overall strategies that L2 learners use. 

Cognitive strategies help learners to identify, change and manipulate the language. Memory strategies are those that help the learner to remember and recall key items. They include semantic mapping and word associations, using keywords and developing mental images of words and phrases though grouping. Compensation strategies with regard to reading strategies, include guessing lexical meanings from the text and making inferences. Metacognitive strategies are strategies employed by learners to self-monitor, plan and execute their own learning. Social strategies include gaining the help of others when learning. This might include asking peers or the teacher for help or asking for feedback and correction. Finally, affective strategies can be identified that include those strategies that help learners to reduce anxiety and encourage their own attitude towards learning. Oxford (1990, p.16) categorises these six strategy sub-groups into two main concepts, namely those that are ‘direct’ strategies (cognitive, memory, compensation) and those that are ‘indirect’ strategies (metacognitive, social, affective). However there is no attempt to indicate how effective each strategy is, and whether they are used alone or in combination of other strategies.

Macaro (2001) offers a clearer way of conceptualising these classifications of learning strategies, suggesting that all types of strategies lie along a continuum, arguing that this representation better accounts for how strategies are classified as there is,…considerable overlap between cognitive and matacognitive strategies,  …usually brought about by the situation in which the strategy is taking place  (Macaro, 2001, p.24). One could further assume that not only do ‘direct’ and ‘indirect’ strategies overlap but that when a learner attempts a task, he/she will combine available (and known) strategies in a manner that affects maximum learning efficiency.

Early L2 reading strategy studies
Early research in L1 and L2 reading ( Brown, 1981; Baker and Brown, 1984; Hosenfeld, 1977) has indicated that using a combination of strategies facilitates more proficient reading comprehension. Hosenfeld (1977) designed a think-aloud protocol to ascertain reading strategies used by proficient L2 readers, and those by non-proficient readers. Results showed that good readers used combinations of strategies: skipping non-essential or less important words; reading longer phrases at a time; keeping the meaning of a text active whilst reading; maintaining a positive attitude towards reading.

Effective strategy use
Hosenfeld`s study, implies that to be able to apply effective combinations of strategies, a learner has to have a good awareness and knowledge of the resources available to them. Baker and Brown (1984) support this theory, relating metacognitive ability to the reader’s cognitive or metacognitive awareness (declarative knowledge) and to their understanding of how and when to use these resources (procedural knowledge). They also relate metacognitive ability to how well a reader knows herself/himself. An element of declarative knowledge includes the level of familiarisation with a text at hand.
PROCEDURAL
KNOWLEDGE
　　　　　　　　　　　　　　　　　　　　

Fig 1: Model of Metacognitive ability related to procedural and declarative knowledge

We can therefore list a number of important qualities related to metacognition that appear to be vital for facilitating  comprehension.
Description of  ‘declarative’ elements:  ‘Self-knowledge’ includes recognising the benefits of positive attitudes towards reading, and the level of one’s own reading ability. Unawareness of personal limitations restricts a reader’s ability to anticipate or control potential difficulties (Carrell, 1989), for example, the level of one’s own performance related to text type/genre [eg. narrative texts (fiction/non-fiction/fable), factual texts (newspaper articles/event programmes), poems, etc.]. ‘Task- knowledge’ includes recognising potential difficulties of a task, for example, that familiar text types or topics are more easily accessed than those less familiar and that unambiguous clauses related to previous clauses, help a reader to build textual meaning.

Description of ‘procedural’ knowledge:‘Procedural’ knowledge includes; recognising how to initially access texts (predictions based on a title, scanning or skimming texts); how to ‘sound out’ lexical items/clauses that help to retain meaning in working memory; and when to activate and use particular strategies necessary for successful comprehension.

Research (Carrell, 1989; Pressley, 1999) has suggested that less successful readers¹ tend to be those who used more local reading strategies, perhaps because they are more dependent on ‘bottom-up’ skills, rather than ‘higher-level’ processing that entails access to background knowledge and making inferences. However, these studies inadequately account for the complex relationships between declarative and procedural elements of metacognitive ability and the successful engagement of texts. One study that stands out however, is a research on the individual differences in reading strategy use for adult second language learners (Anderson, 1991).  Twenty-eight Spanish speaking university students enrolled in an ESL programme were assessed in two types of activities; one whilst taking a reading comprehension test and another while reading academic texts. Prior to the study their general English proficiency level had been assessed through a placement test and their level had been classified as ranging between beginner and advanced level. Think-aloud protocols in response to questions at the end of reading passages were used to evaluate strategy used while reading academic texts and regarding the comprehension tests, questions were grouped according to the type of strategies being evaluated. Results showed from both quantitative and qualitative analysis that the types of cognitive strategies being used to answer questions for both activities were similar across all students. However students whom scored higher marks were judged to be using strategies that were more suitable and effective. This study sheds light on the way good readers not only use combinations of reading strategies to access texts more efficiently and effectively, but also have the knowledge to apply these strategies at the right time and with the correct frequency.

Intervention Study Programmes

L1 and L2 reading intervention studies have generally investigated the effectiveness of various methods of reading strategy scaffolding. In a study of 7th grade students considered to be low-level L1 English readers, Brown  and Palincsar (1984)  investigated the effectiveness of individual strategy training based on four reading strategies; summarizing, questioning, clarifying and predicting. Results indicated statistically significant reading improvement, and adaptations of this study have led research   to  investigate   strategy   training  within  other    learning   environments. 

In a case study researched over two months, English (2003) investigated the effect of reading strategy intervention of five refugee 6th/7th grade students. Research methodology was both qualitative and quantitative in nature. Treatment was in the form of scaffolding adapted from Bernhardt`s (1991) model of successful L2 reading comprehension, viewed as a product of multiple interacting factors, including word recognition, aural/visual decoding, linking textual statements and prior knowledge. Comparing both think aloud protocols and reading comprehension ability tests before and after treatment provided evidence that the participants had statistically improved accessing a variety of text types (‘effect size’ ES= .786).

Song (1998) also provides evidence of statistically improved performance in a university EFL reading classroom. Comparing results based on multiple-choice pre- and post- reading proficiency tests, divided by a 42-hour long strategy training period, Song (1998) concludes that strategy training may be most helpful for weaker readers, and suggests they might have anunawareness, … of the types and value of reading strategies prior to training or might not utilize those strategies even though they may be aware of them (Song, 1998, p.51). Interestingly, although improved scores in general text and inference questions were observed, there was no statistically improved performance regarding detailed critical information.

Finally, in order to support their students` internet and literacy needs, Reinhart and Isbell (2002) developed a model for web-based literacy instruction that introduced critical reading strategies and developed computer literacy skills together in online contexts. Their intervention was focused on current learners using the web, and their study was also used to examine and evaluate online information, much of which is unregulated. Their study, supported by Levene, Ferenz and Reves (2000), provides evidence that a computer based environment combined with scaffolding of reading strategies, helps develop critical EFL reading skills and web-based learner autonomy.

Discussion

Reviewing various models of reading processes and factors affecting L2 literacy, has highlighted the tendency that ‘skilled’ readers possess metacognitive awareness of how, when and with what frequency to use a combination of strategies, while at the same time exhibiting fluent, automatic processing skills. These readers are typically proficient at anticipating potential problems and accessing varying levels of comprehension. Less successful readers in contrast,tend to be those who are unaware of strategy use and use predominantly bottom-up processes. A lack of proceduralisation burdens low-level cognitive processes, to the detriment of skills such as accessing background knowledge and making inferences, often resulting in the breakdown of meaning, which in turn can lead to a decline in motivation and self esteem. These interrelated factors, …are known as the ‘Mathew effect’ ²… and compound the initial reading problems so that they become even more strongly entrenched  (Fawcett and Lynch 2000, p.58). There is therefore strong evidence to indicate that how proficiently students develop strategies for accessing  texts, affects  whether they become  proficient readers or  not.  Recent  studies  on  reading intervention programmes have been encouraging. Scaffolding of successful reading behaviours in the classroom, one-to-one instruction in addition to regular classes and computer assisted reading techniques have highlighted the effectiveness of strategy training in helping learners develop their literacy skills (Leslie and Allen, 1999), ultimately leading towards greater autonomy both inside and outside the classroom.

That many L2 readers at tertiary level are currently at risk of not being able to achieve attainable goals is due in part to the lack of awareness of their own language proficiency skills and low self- esteem. They bring with them various literacy experiences from their previous learning environments as well as different expectations about what they hope, or expect to achieve in an unfamiliar environment.

Authorities ultimately have the responsibility of implementing adequate policies that are specific to this group of learners, and one potential way of supporting these learners is by helping them to evaluate and compensate their strategies for coping with the demands of daily literacy tasks and test taking. This can be achieved in the traditional classroom but also in computer equipped environments that would seem especially appropriate for many Japanese students with limited ‘open-knowledge’ training (McVeigh, 2002).  The question of how best to account for the varied demands of ESL/EFL learners in this environment needs to be researched further, as do issues regarding learner autonomy and teacher training, in order to prevent reading and ultimately academic failure.

_________________________________________
¹ Regarding definitions of ‘successful/good’ and ‘unsuccessful/bad’ readers, care must be taken not to lead towards bias conclusions. These dichotomies are quite simplistic and limiting. Readers cannot naturally be positioned in only one of two distinct categories, and in reality proficiency levels of readers will fall across a broad continuum, ranging from novice to expert.
____________________________________
² The ‘Mathew effect’, a biblical reference to the poor becoming poorer, was described by Stanovich (1986), noting that poor readers become even poorer achievers, eventually leads to school failure.

References
Alderson, J. C. (1984) “ Reading in a foreign language : a reading problem or a language problem?”. In　Alderson J.C and Urquhart A.H. (eds.) Reading in a Foreign Language. London: Longman.
Anderson, N. (1991) Individual differences in strategy use in second language reading and testing. Modern language Journal, 75, 460-472.
Baker, L. and Brown, A. (1984) Metacognitive skills and reading. In D. Pearson (eds) Handbook of reading Research. New York: Longman.
Bartlett, F.C. (1932) Remembering. Cambridge:Cambridge University Press.
Bernhardt, E.B. (1991) Reading development in a second language. Norwood.NJ: Ablex.
Brisbois, J. I (1995) Connections between First- and Second-Language Reading. Journal of Reading Behavior, 27, 4, 565-584.
Brown, A. (1981) Metacognition in reading and writing: The development and facilitation of selective attention strategies for learning from texts. In M. Kamil (eds) Directions in reading research and instruction. Washington D.C, National Reading Conference.
Brown, A. and Palincsar, A. (1984) Reciprocal teaching of comprehension-fostering and comprehension monitoring activities. Cognition and Instruction 1:2 117-175.
Carrell, P. (1989) Metacognitive awareness and second language reading. Modern Language Journal, 73, 121-134.
Chamot, A.U. (1987) The learning strategies of ESL students. In A. Wenden and J. Rubin (eds.) Learner Strategies in Language Learning, Hemel Hempstead: prentice-Hall.
Cumming, A. (1990) Expertise in evaluating second language compositions. Language Testing, 7, 1, 31-51.
Davis, J. N. and Bistodeau, L. (1993) How do L1 and L2 Reading Differ: Evidence from Think Aloud Protocols. Modern Language Journal, 77, 459-472.
English, S. (2003) Towards Effective Reading Comprehension through Strategy Training for EAL Middle School Learners: A Case Study with Intervention. MSc. Thesis, University of Oxford, Department of   Educational Studies Library.
Ellis, G. and Sinclair, B. (1989) Learning to Learn English. Cambridge: Cambridge University Press.
Fawcett, A.J. and Lynch L. (2000) Systematic Identification and Intervention for Reading Difficulty: Case Studies of Children with EAL. Dyslexia, 6, 57-71.
Gillespie, C. (1993) Profile of Adult Learners: Revealing the multiple faces of literacy. TESOL Quarterly, 27, 3, 529-533.
Goodman, K.S. (1970) Reading as a psycholinguistic guessing game. In H. Singer and R.B. Rudell (eds.) Theoretical Models and Processes of Reading. Newark, NJ.: International Reading Association.
Goodman, K.S. (1973) On the psycholinguistic method of teaching reading. In F. Smith (eds.) Psycholinguistics and Reading. New York: Holt, Rinehart and Winston.
Goodman, K.S. (1979) Reading in the bilingual classroom: literacy and biliteracy. Rosslyn, Virginia: National Clearinghouse for Bilingual Education.
Grabe, W. and Stoller, F. (2002) Teaching and Researching Reading. Applied Linguistics in Action Series, Pearson Education.
Hosenfeld, C. (1977) A preliminary investigation of the reading strategies of successful and nonsuccessful second language learners. System 5, 110-123.
Johns, A. (1997) Text, role and context. Cambridge: Cambridge University Press.
Lee , J. and Schallert, D.L. (1997) The Relative Contribution of L2 Language Proficiency and L1 Reading Ability to L2 Reading Performance: A Test of the Threshold Hypothesis in an EFL Context. TESOL Quarterly, 31, 4, 713-739.
Leslie, L., and Allen, L. (1999) Factors that predict success in an early literacy intervention project. Reading Research Quarterly, 34, 4, 404-424.
Levene, A., Ferenz, O. and Reves, T. (2000) EFL Academic Reading and Modern Technology. TESL-EJ 4:4.
Macaro, E. (2001) Learning strategies in foreign and second language classrooms. London: Continuum.
Mackay, R. and Mountford, A. (1979) Reading for information. In R. Mackay, B. Barkman, and R. Jordan (eds.), Reading in a Second language. Rowley, Mass.: Newbury House.
McVeigh, B. (2002) Japanese Higher Education as Myth. M. E. Sharpe Inc. London.
Nix, D. (1983) ‘Links: a teaching approach to developmental progress in children’s reading comprehension and meta-comprehension’. In J. Fine, R. Freedle (eds), Developmental issues in discourse. Norwood, NJ: Ablex.
O’Malley, J. M., and Chamot. A. U. (1990) Learning strategies in second language acquisition. New York: Cambridge University Press.
Oxford, R. (1990) Language learning strategies: What every teacher should know. New York: Newbury House Publishers.
Pressley, M. (1999) Self-regulated compensation processing and its development through instruction. In L.B Gambell, L.M. Morrow, S.B. Neuman, and M. Presseley, Best practices in literacy instruction (90-97). New York: Guilford Press.
Reinhart, J. and Isbell, K. (2002) Building Web Literacy Skills. The Reading Matrix 2:2.
Rogers, K.B.(2002) Re-forming gifted education: Matching the program to the child. Scottsdale AZ: Great Potential Press Inc.
Song, M. (1998) Teaching Reading Strategies in an Ongoing EFL University Reading Classroom. Asian Journal of English Language Teaching Vol. 8: 41-54.
Stanovich, K.E. (1986) Mathew effects in reading: some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360-407.
Wallace, C. (1986) Learning to read in a multicultural society. Oxford: Pergamon.
Zwann, R. and Brown, C.M. (1996) The Influence of Language Proficiency and Comprehension Skill on Situational-Model Construction. Discourse Processes, 21, 289-327.