๐ Lecture 19 โ Teaching Reading: Individual and Social Perspective II
๐ Overview: This lecture examines models of the reading process, from bottom-up and top-down processing to interactive models. It then expands to consider literacy from a broad social perspective, contrasting autonomous and ideological models of literacy, and explores implications for teaching reading in EFL contexts.
๐๏ธ Topics Covered
The lecture covers data-driven models (bottom-up) of reading, concept-driven models (top-down), and interactive models that combine both approaches. It then examines reading from a broad perspective, contrasting autonomous and ideological literacy models, their relationship to cognitive development and economic growth, literacy and reading in specific communities, critical reading approaches, and finally implications for teaching reading through social literacy frameworks including the New London Group's pedagogy and Kern's application to L2/FL learning.
๐ Lecture Summary
Topic-105: Data Driven Models
The bottom-up model of reading (Gough 1972) proposes that the reader processes data from the page in a sequential, letter-by-letter and word-by-word analysis of orthographic words. The processing moves in one directionโfrom the "bottom" (perception of letters) to the "top" (construction of meaning)โand higher-level processing does not affect lower-level processing. In pedagogy, this model justified a phonics-based approach to initial reading that stressed letter-by-letter "sounding out" and included decontextualised exercises like distinguishing minimal pairs (e.g., "park/bark", "tap/top"). The main criticism of data-driven models is that they cannot account for context effects. For instance, initial readers reading in their L1 make miscues (mistakes or deviations from the printed text) that appear generated by their language knowledge, such as an English native-speaker child reading "Rabbit went" for "Rabbit won't" or "He won't bother about..." instead of "He won't bother today..."
Topic-106: Concept-Driven Models
Goodman's psycholinguistic approach to reading can be seen as a reaction against phonics-based methods rather than specifically against Gough's bottom-up model. Proponents of concept-driven (or "top-down") models hold that text is sampled and predictions meaningful to the reader are made based on prior knowledge, especially language knowledge. This is why Goodman called reading a "psycholinguistic guessing game." Although Goodman's account lacks detail compared to Gough's, this view of reading as "guessing" based on the reader's state of knowledge does account for context effects common when initial readers read aloud in their L1. The model exerted considerable influence in applied linguistics and the teaching of initial reading in the USA and UK, particularly through the support of Smith (1978).
Topic-107: Interactive Models
The interactive model was first elaborated by Rumelhart (1977). It proposes that graphemic input (marks on the page) passes to a visual information store where "critical features" are extracted. The extracted information is then operated upon by what the reader knows about languageโsyntactic knowledge, semantic knowledge, lexical knowledge, orthographic knowledge as well as pragmatic information "about the current contextual situation." The crucial point is that these knowledge sources operate in parallel: the information in the pattern synthesiser is scanned to yield the "most probable interpretation," and higher-level processing of meaning may affect lower-level processing (i.e., there is both "top-down" and "bottom-up" processing). The compensatory interactive model (Stanovich 1980) similarly represents reading as involving interaction between bottom-up and top-down processing. The compensatory element claims that a reader's lack of ability at one level may be compensated for by proficiency at anotherโa reader may compensate for weakness at word meaning level by drawing on appropriate background knowledge. There are clear advantages of such a view for L2 reading.
Topic-108: Reading: The Broad Perspective
Reading in the broad perspective is concerned not with psycholinguistic processes or comprehension levels, but with literacy as social practiceโsocial patterns of activities involving reading and writing, as well as the social values attached to these activities. An important distinction is between the "autonomous literacy" model and the "ideological literacy" model (Street 1984). The autonomous model sees literacy as a value-neutral set of skills detached from social context, whose possession is assumed to bring certain cognitive and social results. Much of what has been described as the "narrow" approach to literacy is in this tradition. In contrast, the ideological model is concerned with literacy practices in relation to specific social contexts; the multiplicity of contexts generates a multiplicity of literacies, which are associated with power and ideology. The ideological model leads to a better understanding of how literacy is embedded in other human activityโ"literacy" does not exist outside of human action, and the strong may manipulate institutions concerned with literacy in ways that disadvantage the weak.
Supporters of the ideological model claim that two invalid claims are made for autonomous literacy: 1) that literacy leads to logical and scientific thinking, and 2) that literacy leads to social and economic development. The first claim (made by anthropologist Goody) is challenged by Scribner and Cole's (1981) research on the Vai people in Liberia, where one group was literate in the Vai script, another had literacy in reading the Koran, and a third was literate in English (the medium of education). Their conclusions are frequently cited to claim that it is not literacy itself that produces cognitive changes, but schooling, since the schooled group literate in English were superior in reasoning power.
Regarding literacy and economic development, Denison (1962) claimed that between 1930 and 1960, 23% of annual growth in US national income could be attributed to education. Anderson (1966) estimated that an adult literacy rate of about 40% was needed for economic development, though he added this level would not be sufficient if societies lacked other support systems. The failure of the Experimental World Literacy Programme (organized by UNESCO in 11 countries from 1967 to 1972) to generate economic growth proved that literacy alone cannot be a causal factor in development. UNESCO concluded that literacy programmes should be integrated with economic and social reforms. However, Azariadis and Drazen (1990), studying 32 countries from 1940 to 1980, concluded that none of the countries where the level of education (including literacy) was inadequate managed to achieve rapid growth. The current consensus is that literacy is a necessary contributory factor in development, but not an independent causal factor.
Topic-109: Literacy and Reading
The new literacy studies have carried out detailed ethnographic work on reading and writing practices in specific communities. Examples include Heath's (1983) seminal work on literacy in three US communities, Barton and Hamilton's (1998) description of literacy practices in Lancaster, and Martin-Jones and Jones's (2000) documentation of bilingual literacies. Street's (1984) research on literacy in Iranian villages identified three sets of literacy practices: traditional literacy associated with primary Quranic school; schooled literacy from modern state school; and commercial literacy associated with selling fruit. Contrary to expectation, commercial literacy was mainly undertaken by those who had Quranic literacy, since they had the social status within the village that people with only schooled literacy lacked.
A second concern of the broad approach is critical reading, deriving from critical discourse analysis, which attempts not only to describe texts but also to interpret and explain them. Critical readings typically examine: 1) linguistic issues (choice of vocabulary, manipulation of grammar such as expression or suppression of agency in verb phrases); 2) rhetorical issues (overall text structure and organisation); and 3) issues of text type and discourse convention. The approach may critique not only the language and sentiments expressed but also the ideological and/or historical assumptions underpinning them. Critical reading claims to "look beyond the classroom to the way in which reading practices are carried out and perceived in the wider society" (Wallace 1996: 83). While probably not suited to low-level EFL learners, it is claimed to be both possible and desirable for learners with adequate English.
The broad approach presents a strong moral argument, but its proponents occasionally provide incomplete representations of the psycholinguistic tradition. Gee (1996) claimed that the psycholinguistic position holds there is a "right" interpretation for texts that is roughly the same for all competent readersโa notion widely disputed by applied linguists. However, if one cannot read in the psycholinguistic sense, one cannot make any interpretation of any written text. Thus, the autonomous model is valid in that if one cannot read, one cannot read anything, while the ideological model is valid in that the converse proposition "If one can read, then one can read everything" is incorrect.
One of the chief merits of the new literacy studies is focusing attention on the social dimension and making the point that literacy practices are ideologically laden and often manipulated by powerful institutions. Gregory and Williams (2000) documented home and school practices in a multicultural urban area of London, finding that children from economically poor backgrounds draw on home literacy practices as well as school practices in learning to read, with older siblings and grandparents as important mediators of literacy.
Topic-110: Literacy and Implications for Teaching Reading
A proposal for implementing a pedagogy drawn from social literacies has come from the New London Group. Having developed the basic concept of "Design" (referring to conventions of meaningโlinguistic, visual, audio, gestural, and spatial), the group proposes four sequential components of pedagogy:
- Situated Practice: draws on students' experience of meaning-making in their lives
- Overt Instruction: through which students develop an explicit metalanguage of Design
- Critical Framing: interprets the social context and purpose of Designs of meaning
- Transformed Practice: in which students become "designers of social futures"
Kern (2000: 129-170) applies these four components to reading in L2/FL contexts. For Kern, "situated practice" is largely student-centered activity with group predictions and negotiations about text meaning; overt instruction consists of work on lexical choices, syntactic relations, and discourse structure; critical framing involves students distancing themselves from the text through critical questioning and summarising work; transformed practice is essentially a matter of writing, with Kern suggesting translation and transforming a text into a dialogue. What the approach stresses is the critical perspectiveโcomparing and discussing interpretations of students and teachers rather than extracting fixed meanings, and encouraging students to be aware of the social context in which the text was produced as well as the social context in which they as L2 readers are interpreting it.
There are relatively few practical examples of EFL work in this framework, possibly because the approach has no obvious theory of learning. Street (2003: 85) suggests the emphasis should be on "appropriateness," implying students should explore the various uses and meanings of literacy in the social context of the school and its surrounding communities, advocating a "combination of ethnographic-style research into everyday literacy practices and constructive curriculum development and pedagogy."
โญ Key Takeaways
The lecture presents three models of the reading processโbottom-up (data-driven), top-down (concept-driven), and interactive/compensatory modelsโwith interactive models best accounting for how readers use multiple knowledge sources simultaneously. In the broad perspective, literacy is understood as social practice rather than merely a cognitive skill, with the autonomous model viewing literacy as value-neutral skills and the ideological model viewing literacy as embedded in power relations and social contexts. Evidence suggests literacy alone does not produce cognitive changes (schooling does) nor independent economic development (it is necessary but not sufficient). The New London Group's pedagogy of Design (Situated Practice, Overt Instruction, Critical Framing, Transformed Practice) and Kern's application to reading offer frameworks for implementing social literacy approaches in EFL teaching, though practical examples remain limited.
๐ง Quick Revision Questions
- What are the three main models of the reading process discussed in this lecture, and how does the interactive model differ from both bottom-up and top-down models?
- What evidence from the Vai people in Liberia challenged the claim that literacy leads to logical and scientific thinking?
- What did the failure of the UNESCO Experimental World Literacy Programme (1967-1972) demonstrate about the relationship between literacy and economic development?
- What are the four sequential components of pedagogy proposed by the New London Group, and how does Kern apply them to L2/FL reading instruction?
- According to Street's research on Iranian villages, why did people with Quranic literacy rather than schooled literacy primarily undertake commercial literacy?
๐ Lecture 20 โ Research Informing L2 Reading I
๐ Overview: This lecture examines how research on first language (L1) and second language (L2) reading informs instructional practices. It establishes that reading is a complex cognitive skill requiring explicit instruction, and explores nine key implications for effective L2 reading instruction, focusing on the first five: word recognition fluency, vocabulary learning, background knowledge activation, grammar knowledge, and comprehension strategies.
๐๏ธ Topics Covered
The lecture begins by introducing reading as a complex cognitive activity that is not naturally acquired like speaking, then presents nine research-supported implications for L2 reading instruction. It covers word recognition fluency research demonstrating automaticity's importance, vocabulary-rich environments and their strong correlations with comprehension, activation of background knowledge through text previewing, the significant role of grammar knowledge in reading ability, and specific comprehension strategies validated by the National Reading Panel.
๐ Lecture Summary
Topic-111: Introduction to Research Informing L2 reading
Reading is a complex cognitive activity involving the secondary uses of cognitive skills in relatively new ways. Unlike speaking and listening in a first language (L1), reading is not an inherently natural process. Learning to read requires considerable cognitive effort and a long learning process, whether in L1 or L2. If a person is not taught to read, they will not learn to read.
Variables affecting reading instruction include student proficiency, age, L1/L2 relations, motivation, cognitive processing factors, teacher factors, curriculum and materials resources, instructional setting, and institutional factors. Despite this complexity, many learners become fluent L2 readers.
Three reasons for optimism in connecting research to practice:
- L1 reading research has made remarkable advances in the past 15 years
- L2 reading research provides additional insights converging with L1 findings
- Real distinctions between L1 and L2 reading do not prevent drawing implications from L1 research
๐ก Why this matters: Understanding that reading must be explicitly taughtโunlike oral languageโshapes how teachers approach L2 reading instruction from the beginning.
This overview focuses on learners developing academic reading abilities in school settings. The purpose is to link research findings to key implications for instruction. The review will not separate L1 from L2 research but will refer specifically to L2 research when recent studies apply to instructional practices.
Topic-112: Implications for Reading Instruction from Reading Research
Nine research-supported implications for academic reading instruction and curriculum design have emerged:
- Ensure word recognition fluency
- Emphasize vocabulary learning and create a vocabulary-rich environment
- Activate background knowledge in appropriate ways
- Ensure effective language knowledge and general comprehension skills
- Teach text structures and discourse organization
- Promote the strategic reader rather than teach individual strategies
- Build reading fluency and rate
- Promote extensive reading
- Develop intrinsic motivation for reading
A long list of implications does not represent a ready-made curriculum. Any instructional setting must determine priorities based on student needs, institutional expectations, and resource constraints. The choices of which factors to emphasize rest with local contexts and goals.
Topic-113: Ensure Word Recognition Fluency
Word recognition fluency has been widely recognized in L1 reading research as an important factor in explaining reading comprehension abilities, particularly at earlier stages of reading development. Segalowitz (1991) demonstrated that word recognition automaticity was an important factor in distinguishing proficiency levels of very advanced L2 readers. Segalowitz, Segalowitz, and Wood (1998) showed that L2 university students who were more fluent readers overall had better word recognition automaticity skills. Less fluent students improved their L2 word recognition automaticity through instruction over an academic year. Fukkink, Hulstijn, and Simis (2003) reported fluency gains through word recognition training for eighth grade EFL students in Holland with just two training sessions.
The second issue is whether fluency can be taught and whether it improves comprehension. Repeated exposures to high frequency words through extensive reading contribute to automatic word recognition, but no causal connection has been demonstrated in L2 settings. In L1 research, Tan and Nicholson (1997) trained below-average grade 3-5 students using flash card practice; experimental students outperformed controls in both fluency and passage comprehension. Levy, Abello, and Lysynchuk (1997) demonstrated that both word recognition training and repeated readings positively impacted comprehension of texts containing trained words.
A final issue involves how to teach word recognition fluency effectively through timed word recognition practice, greater phonological awareness, morphological awareness training, extended reading practice, and assisted reading activities.
Topic-114: Creating Vocabulary Rich Environment
The relation between vocabulary knowledge and reading comprehension has been powerfully demonstrated in both L1 and L2 contexts. Anyone who wants to be a fluent reader must have a large vocabulary.
In L1 research, Thorndike (1973) surveyed reading in 15 countries with over 100,000 students and reported median correlations between reading and vocabulary of r = .66 to r = .75. Carver (2003) argued that when reliable vocabulary tests and reading comprehension measures are converted to grade-level equivalent scores, corrected correlations will be almost perfect.
In L2 settings:
- Droop and Verhoeven (2003): demonstrated powerful relation between vocabulary knowledge and later reading ability with 3rd and 4th grade language minority children in Holland
- Pike (1979): reported corrected correlations between vocabulary and reading on TOEFL of .84 to .95
- Laufer (1997): cited correlations of .50 to .75
- Qian (2002): found correlations from .68 to .82 between TOEFL reading and vocabulary measures
Topic-115: Activate Background Knowledge in Appropriate Ways
Almost all reading researchers agree that background knowledge plays an important role in reading comprehension. Readers comprehend texts better when texts are culturally familiar or relate to well-developed disciplinary knowledge. Background knowledge is essential for inferences, text model construction, disambiguating lexical meanings, and syntactic ambiguities.
Complications arise with texts presenting new information. The limited role of background knowledge for comprehending new topics was documented by Bernhardt (1991). Alderson (2000) presents conflicting evidence on background knowledge's role in reading assessment.
From an instructional perspective, Chen and Graves (1995) demonstrated that text previewing led to significantly better comprehension compared to both a control group and a group activating general background knowledge. This supports activating specific information relevant to the text rather than general background knowledge.
Topic-116: Language Knowledge and General Comprehension Skills
Text comprehension requires both a) language knowledge (vocabulary and grammar) and b) recognition of key ideas and their relationships through comprehension strategies.
Research shows strong relations between grammar knowledge and reading:
- Van Gelderen et al. (2002): correlation of .73 between EFL L2 grammar and reading; .78 for EFL L3
- Alderson (1993): correlations of .80
- Pike (1979): corrected correlations of .80 to .85 among TOEFL subsections
- Enright et al. (2002): correlation of .91 between structure and reading on current TOEFL; .83 between structure and piloted New TOEFL reading
However, this has not led to a call for extended grammar instruction as direct support for L2 reading. At advanced levels, grammar is better seen as an indirect support system developed through comprehension instruction and strategy training.
The National Reading Panel (2000) and Trabasso and Bouchard (2002) identified nine individual reading strategies with significant influence on reading comprehension:
- Prior knowledge activation
- Mental imagery
- Graphic organizers
- Text structure awareness
- Comprehension monitoring
- Question answering
- Question generating
- Mnemonic support practice
- Summarization
There is relatively little recent L2 research on specific comprehension strategies, but it is reasonable to extrapolate from L1 research that these strategies are useful for developing student reading comprehension.
โญ Key Takeaways
Students must remember that reading is a taught skill, not a natural acquisition, and that nine research-supported implications guide effective L2 reading instruction. Word recognition fluency and automaticity are foundational, with studies showing that even brief training can improve fluency. A powerful, near-perfect relationship exists between vocabulary knowledge and reading comprehension across both L1 and L2 contexts. Background knowledge should be activated through specific text previewing rather than general preparation. Finally, grammar knowledge strongly correlates with reading ability, but it functions best as an indirect support developed through comprehension strategy instruction rather than isolated grammar teaching.
๐ง Quick Revision Questions
- Why is reading considered "not an inherently natural process" unlike speaking and listening?
- What did Segalowitz, Segalowitz, and Wood (1998) demonstrate about L2 word recognition automaticity and instruction?
- What correlation ranges did Thorndike (1973) and Qian (2002) find between vocabulary and reading comprehension?
- According to Chen and Graves (1995), what type of background knowledge activation led to significantly better comprehension?
- What nine individual reading strategies did the National Reading Panel identify as having significant influence on reading comprehension?
๐ Lecture 21 โ Research Informing L2 Reading II
๐ Overview: This lecture explores how research evidence from both L1 and L2 settings informs effective reading instruction practices. It covers five key research-informed approaches: teaching text structure, promoting strategic reading, building reading fluency, encouraging extensive reading, and developing intrinsic motivation for reading. Understanding these research foundations is critical for designing effective L2 reading curricula.
๐๏ธ Topics Covered
The lecture examines five major research areas that inform L2 reading instruction: teaching text structure and discourse organization through three lines of research; promoting strategic readers through coordinated multi-strategy instruction approaches like TSI and CORI; building reading fluency through repeated reading programs; promoting extensive reading as a support for comprehension and vocabulary development; and developing intrinsic motivation for reading, particularly through Concept-Oriented Reading Instruction.
๐ Lecture Summary
Topic-117: Teach Text Structure and Discourse Organization
For older and more advanced L2 students, instruction emphasizes expository prose processing for learning purposes. Students must understand the abstract patterns of text structuring in expository prose that support comprehension. Texts incorporate discourse structures (also called knowledge structures or basic rhetorical patterns) that have functional purposes such as comparing two ideas or highlighting cause-and-effect relationships. Good readers and writers recognize these purposes, often implicitly. These discourse mechanisms extend to the level of genre and larger frames of discourse structure.
๐ Definition โ Discourse structures: Functional organizational patterns in texts (e.g., compare/contrast, cause/effect) that guide readers toward preferred interpretations.
Three major lines of research (mostly L1) examine the effect of text structure instruction:
- Direct instruction explicitly raising student awareness of specific text structuring. Meyer and Poon (2001) demonstrated that structure strategy training significantly improved recall for both younger and older adults.
- Graphic organizers, semantic maps, outline grids, tree diagrams, and hierarchical summaries (Tang 1992; Trabasso and Bouchard 2002). Students comprehend texts better when shown visually how information is organized.
- Reading strategy training including summarizing, semantic mapping, predicting, forming questions from headings/sub-headings, and using adjunct questions improves awareness of text structure and comprehension (Duke and Pearson 2002; Trabasso and Bouchard 2002).
๐ก Why this matters: While L1 research demonstrates clear benefits of text structure instruction, relatively little L2 research exists in this area, though L1 findings likely apply well to L2 students.
Topic-118: Promote the Strategic Reader rather than Teach Individual Strategies
L1 reading comprehension instruction today is equated with strategic reading development. Research shows that comprehension is strongly influenced by instruction emphasizing the coordinated use of multiple strategies. Such instruction involves direct teaching of several strategies while students read and comprehend a text. The teacher and students engage in discussions about the text while learning to use key strategies in effective combinations. Students learn through teacher modeling, teacher scaffolding and support, and gradual independent use of strategies.
Most multi-strategy approaches focus on 4-8 major strategies, though some incorporate up to 20-30 distinct strategies over longer periods. Two proven L1 approaches applicable to L2 settings:
- Transactional Strategies Instruction (TSI) โ Provides a curricular framework for strategic comprehension instruction, incorporating activities beyond strategy development (vocabulary, fluency, extensive reading).
- Concept-Oriented Reading Instruction (CORI) โ Fully engages students in all aspects of strategic reading instruction (Guthrie et al. 1999; Guthrie, Wigfield, and von Secker 2000).
L2 research has not developed as extensively in strategic engagement frameworks, but notable approaches include: Janzen's (2001) L2 adaptation of TSI, Klingler and Vaughn's (2000) Collaborative Strategies Instruction, Cognitive Academic Language Learning Approach (CALLA) (Chamot and O'Malley 1994), and Strategy-Based Instruction (SBI) (Cohen 1998).
Most L1 researchers now see value in teaching strategies as combined-strategies instruction rather than as independent processes taught separately from basic comprehension.
Topic-119: Building Reading Fluency and Rate
Reading fluency has gained much greater importance in recent years, particularly in L1 settings.
๐ Definition โ Reading fluency: Involving both word recognition accuracy and automaticity; requiring rapid speed of processing across extended text (reading efficiency); making appropriate use of prosodic and syntactic structures; sustainable for extended periods; taking a long time to develop (National Reading Panel 2000; Segalowitz 2000; Kuhn and Stahl 2003).
The National Reading Panel (2000) devoted a major section to fluency research, demonstrating through meta-analysis that fluency can be taught and positively impacts reading comprehension. Kuhn and Stahl (2003) reached similar conclusions with a more inclusive meta-analysis. In L1 settings, almost any kind of independent or assisted repeated reading program, done carefully, has a direct positive effect on reading fluency and an indirect positive effect on comprehension improvement.
L2 research on fluency training is limited but emerging. The best ongoing exploration is Taguchi's work (1999; Taguchi and Gorsuch 2002). Both studies showed that repeated reading of short graded readers leads to improved reading fluency. The more recent study demonstrated that students read significantly faster in post-reading tests than pre-reading tests while maintaining the same levels of comprehension.
Topic-120: Promote Extensive Reading
The true experimental research on extensive reading is seemingly contradictory, but the preponderance of non-experimental research overwhelmingly supports extensive reading for reading comprehension development, reading fluency, incidental vocabulary learning, and word recognition fluency.
The National Reading Panel (2000) found no single experimental study demonstrating significantly better reading comprehension for an extensive reading group. However, Kuhn and Stahl (2003) pointed out that the restricted range of studies reviewed ruled out much persuasive research.
Two specific research programs strongly support extensive reading:
- Stanovich (2000) and colleagues demonstrated across multiple studies (1990-2000) that the amount of overall exposure to print by readers has a direct relation to vocabulary knowledge and comprehension abilities.
- Guthrie et al. (1999) demonstrated that for students from grades 3 to 10, amount of reading significantly predicted text comprehension.
In L2 settings, Elley (2000) provides the strongest ongoing evidence, reviewing book flood approaches that include additional instructional practices. Reporting on large-scale curricular research studies (Niue, Fiji, Singapore, Sri Lanka, South Africa, Solomon Islands, 1977-1998), he demonstrated that modified book floodsโalong with careful teacher trainingโlead consistently to significant results in comprehension development.
Topic-121: Develop Intrinsic Motivation for Reading
In L1 settings, the strongest evidence of motivation's direct impact on reading comes from Guthrie and colleagues:
- Wigfield and Guthrie (1997) demonstrated that motivation and engagement with reading were significantly related to amount of reading. More highly motivated fourth and fifth grade students engaged in significantly more reading.
- Guthrie et al. (1999) demonstrated that higher motivation among third and fifth grade students significantly increased their amount of reading and text comprehension.
- Guthrie et al. (1998) demonstrated that Concept-Oriented Reading Instruction (CORI) developed significantly higher levels of student motivation than control classes among third and fifth grade students.
In L2 settings, there is little research specifically on the relation between motivational variables and reading comprehension. Most L2 motivation research focuses more generally on language abilities. Dรถrnyei (2001) provides an excellent overview of motivational factors and their influences on L2 learning, covering both research and attention to motivation instruction and teacher motivation.
โญ Key Takeaways
The lecture establishes five research-informed pillars for effective L2 reading instruction that students must remember for exams. First, teaching text structure through direct instruction, graphic organizers, and strategy training significantly improves comprehension, though more L2-specific research is needed. Second, promoting strategic readers through coordinated multi-strategy instruction (like TSI and CORI) is more effective than teaching individual strategies in isolation. Third, reading fluency can be directly taught through repeated reading programs, which positively impact both fluency and comprehension. Fourth, extensive reading is strongly supported by non-experimental research for developing comprehension, fluency, and vocabulary, with book flood approaches showing particular promise in L2 settings. Finally, intrinsic motivation for reading can be developed through instructional approaches like CORI, and motivation directly predicts reading amount and comprehension outcomes.
๐ง Quick Revision Questions
-
What are the three major lines of L1 research on teaching text structure, and how do they support reading comprehension?
-
Why is combined-strategies instruction considered more effective than teaching individual reading strategies separately?
-
What is the definition of reading fluency, and what evidence supports the effectiveness of repeated reading programs in L2 settings?
-
What does the National Reading Panel's meta-analysis say about extensive reading, and what counter-evidence do Kuhn and Stahl provide?
-
How did Guthrie and colleagues demonstrate that intrinsic motivation for reading can be directly taught through classroom instruction?
๐ Lecture 22 โ ASSESSING READING I
๐ Overview: This lecture introduces the fundamental reasons and principles for assessing reading skills, including motivation, achievement measurement, and diagnosis. It covers different assessment formats such as pronominal questions, multiple-choice questions, and information transfer, while emphasizing the importance of reliability, validity, and practicality. Understanding these assessment types is essential for educators to effectively evaluate and support reading development.
๐๏ธ Topics Covered
The lecture begins with an introduction to assessing reading, covering reasons such as encouraging learning, monitoring progress, diagnosing problems, and measuring proficiency. It then explores motivation as an informal assessment tool. The focus shifts to measuring achievement, examining various comprehension test formats including pronominal questions/imperatives, multiple-choice questions, and information transfer. Finally, the lecture addresses diagnosing reading problems, outlining four general principles for effective diagnostic testing.
๐ Lecture Summary
Topic-122: Introduction to Assessing Reading
There are several reasons for assessing reading and the skills and knowledge involved in reading. These include assessing to encourage learning, assessing to monitor progress and provide feedback, assessing to diagnose problems, and assessing to measure proficiency. The same form of assessment may be used for a variety of goals. Good assessment needs to be reliable, valid, and practical. Reliability is helped by having a high number of points of measurement, by using a test format learners are familiar with, and by using consistent delivery and marking procedures. Validity is helped by using reliable measures, and by being clear about what is being measured and why. The practicality of a test can be helped by giving very careful thought to how learners will answer the test and how it will be marked. The ease of making a test is also part of its practicality.
Topic-123: Motivation as an Informal Assessment
A very common use of informal assessment is to make learners study. At worst, they study because there will be a test, but preferably success in the test maintains their interest in study. Regular comprehension tests can do this, but there are other ways as well which do not involve formal testing.
Topic-124: Measuring Achievement
Measures of achievement focus on the learning done in a particular course. If a course has focused on speed reading, then the achievement measure would be a speed reading measure even though speed of reading is only a part of the larger picture of reading proficiency. Achievement measures are thus closely related to the course of which they are part. They need to have a high level of face validity; that is, they should clearly look like what they are supposed to be measuring. Achievement tests could test various reading strategies, speed of reading, word recognition, reading aloud, or note-taking from reading, depending on the goals of the course.
๐ Definition โ Face Validity: The extent to which a test appears to measure what it is supposed to measure, based on superficial inspection.
Pronominal Questions, Imperatives These questions require learners to make a written answer ranging from a single word to several paragraphs. Usually for comprehension, short answers are required and these are called short answer questions. If answers are short, more questions can be answered, increasing reliability and validity. These questions can be used for all focuses of comprehension. They are suited to checking literal comprehension because it is not difficult to write questions avoiding same words used in the text. They are suited to inferences, application and responding critically because learners must search for and construct their own answers. A positive feature is that they can be marked using a grading scale, for example 0, ยฝ, 1 or 0, 1, 2, 3 marks for each question depending on completeness and accuracy, allowing credit for partial comprehension.
๐ก Why this matters: Short answer questions allow for nuanced assessment of comprehension depth, but introduce reliability challenges due to varied learner responses.
The disadvantages include the problem of consistency in marking when learners write a range of differently worded answers. This is primarily an issue of reliability. It is best dealt with by making and adding to a list of possible answers with corresponding marks, and getting another teacher or highly proficient reader to answer the test and then check the list of possible answers. This prechecking may result in changes to questions to limit possible answers. The answer sheet can have a set space for each answer.
Topic-125: Multiple Choice Questions
These question forms are grouped together because the answer is contained within the question or instructions, so learners do not have to compose their answer, which simplifies marking. Typically there is a stem with four choices, one of which is correct. To produce a large number of questions for reliability, a long text or several short texts are needed. Marking is usually very easy, and most learners are familiar with multiple-choice tests. Good multiple-choice tests tend to be very reliable.
Multiple-choice questions can focus on details (microstructure) and on more general aspects (macrostructure) of the text, although some researchers have found difficulty in using multiple-choice to measure global comprehension. Multiple-choice tests only involve reading, so measurement is less likely to be affected by writing skill than in short answer tests. If a test is not well prepared, learners may get a reasonable score without reading the text. Checking involves ensuring the correct answer is not always shorter or longer than the distractors, asking a native speaker to answer the test, getting a colleague to critically review items, and looking at learners' answers to see if some items are too easy or too hard.
๐ Definition โ Stem: The part of a multiple-choice question that presents the problem or question to be answered. ๐ Definition โ Distractors: The incorrect answer choices in a multiple-choice question, designed to be plausible but wrong.
To make marking easier, a special answer sheet and an answer key may be used. Learners circle the correct answer. Because of ease of marking, multiple-choice is useful for very large numbers of tests, and computer marking is possible. Practicality is a strength of using multiple-choice tests, but making them is not easy; creating four plausible choices is usually a challenge and good questions require a lot of trialling.
Information Transfer Incomplete information transfer diagrams can be used to measure comprehension of a text. Learners read the text and fill in the diagram with short notes. The advantages are that the information produced can cover many points without a lot of writing. The disadvantage is in gaining consistency in marking.
Topic-126: Diagnosing Problems
If a learner has problems with reading, it is useful to see where the problems lie. As reading is a complex skill, there are many possible sources of difficulty. If a learner performs badly on a proficiency measure such as a cloze test or a comprehension test, a procedure is needed to find reasons for poor performance.
There are four general principles to follow:
- Diagnosing problems should be done on an individual basis โ with the teacher sitting next to the individual learner and carefully observing. If testing is done with the whole class, individual learners may not give their best effort. A teacher needs to observe what aspects are causing difficulty and adjust the procedure during the process.
- Begin with the smallest units and go step by step to larger units โ starting with word identification, moving to vocabulary knowledge, then to comprehension of single sentences, and then to text comprehension and reading speed. The assumption is that smaller units combine to contribute to larger units.
- Learners should feel comfortable and relaxed during diagnostic testing โ this can be applied by beginning with very easy tasks where the learner can be successful, giving praise for effort and success, being friendly, and taking small breaks.
- Do not rely on only one test โ even where the problem seems obvious, use a different kind of test possibly at a different level of unit size to double check. Decisions about a learner's skill level can have far-reaching effects on their learning.
โญ Key Takeaways
The lecture establishes that effective reading assessment must balance reliability, validity, and practicality, with different assessment formats serving distinct purposes. Short answer questions allow for detailed comprehension checking including inferences and critical responses but require careful marking procedures to maintain reliability. Multiple-choice questions offer high practicality and reliability for large-scale testing but are difficult to construct and may not measure global comprehension effectively. When diagnosing reading problems, teachers must follow a systematic individual approach starting with the smallest units (word identification) and progressing to larger ones, while ensuring the learner remains comfortable and using multiple tests for verification.
๐ง Quick Revision Questions
- What are the four main reasons for assessing reading mentioned in the lecture?
- How do short answer questions help assess inferences and critical responses, and what is their main disadvantage?
- What are the three criteria that good assessment needs to meet, and what factors contribute to each?
- What are the four general principles for diagnosing reading problems, and why is individual testing emphasized?
- Why might multiple-choice questions be less effective for measuring global comprehension, and what steps can be taken to improve their validity?