How I adapt content for ML / ELL students
Challenges that Multilingual Students (MLs) Encounter
Some challenges for ML students may feel obvious, but there are many little hurdles that they can encounter within a seemingly simple lesson. However, depending on their English proficiency, anything can be a roadblock preventing students from understanding the task, doing the intellectual work, and then showing their learning. Let’s focus for a minute on the Identifying Elements Based on Spectra activity in the first lesson for my unit. That is a good example where explaining to a student what they are supposed to do might be the hardest part. The vast majority of students can understand the task on an intellectual level: use the glasses to look at the bulb, compare what I see to the sheet with example pictures, pick the one I think is most alike, then write the name of the one I pick on the blank space. Even if they cannot read English, students can still copy the name onto their page. However, communicating to a student who speaks and reads little English what you want them to do could be difficult. If your student cannot speak nor read then interacting becomes slow, intentional, and multi-faceted with speaking, gesturing, and drawing. As this example illustrates, completing simple tasks can be hard for multilingual (ML) students, but the example, importantly, also illustrates that even a student who speaks absolutely zero English can participate in the spectra activity and complete it once you help them understand the directions. So try to think of ML struggles as metaphorical hurdles that they need help jumping over before they continue participating, not dead ends where there is no room for progress. Your students can and want to participate in class, they just need extra assistance of some form to be able to access the content and show their learning.
Some of the hurdles that I find Multilingual (ML) and English Language Learner (ELL) students run into most often are:
They need longer think time to speak and write
Every time they communicate, they need to translate the question into their first language (L1), decide what the answer is, and then translate their answer back into English. So one step for English speakers becomes three steps for ML students. Teachers and group mates need to wait and give them time to think. Sometimes a student needs 10 seconds, 30 seconds, a minute. And while it feels like a very long time in the moment, you need to wait until they respond either with an answer or an “I don’t know”.
This is also why I try to pair beginning ML students with more patient English speakers, so that their English speaking group members are less likely to get frustrated with MLs and ignore them or leave them behind.
Listening is an active task that requires effort and time to process
Your multilingual students are actively translating and decoding everything you say to understand you. It’s important to try and give them “Comprehensible Input”. Comprehensible input is the multifaceted communication I hinted to in the example above. It is all the things you can do to make your communication more understandable: using pictures and diagrams to help show meaning, pointing towards what you are talking about, acting out actions, writing examples or notes on the board, drawing pictures to show what you are saying, modifying text complexity and difficulty to meet a student’s abilities, and definitely speaking intentionally. When speaking you want to speak slowly and clearly because your student has to think about each word. Use correct English grammar, but use shorter simpler sentence structure; instead of a compound sentence like this one you would probably separate it into two small simple sentences. Pause between each sentence or between phrases. Earlier in this paragraph when I listed all the things to be more comprehensible, I would have deliberately paused at the end of each idea.
I first encountered the idea of Comprehensible Input from an article written by Susan Gass. “One thing that is clear is that the language spoken to non-proficient learners differs from that spoken to proficient speakers (either native or non-native speakers). In general, speech is often slower, simpler in grammatical structures, simpler in vocabulary, and simpler in delivery (e.g., contractions are eliminated).”[1] However, I have expanded my personal understanding to be more than just how you speak to a student. I think about it as all the ways to communicate to a student both verbal and non-verbal.
Content Vocabulary and Academic English need to be intentionally introduced and modelled
Students will learn conversational English much faster than vocabulary that is related to a specific subject or academic English (which is non-specific but high tier English words). If you think about it, multilingual students will hear words like bathroom, desk, lunch, hour in all their classes, but they will probably only hear and use “absorption spectrum” in science class. So these kinds of words need to be deliberately introduced by the teacher, defined, accompanied by example pictures or diagrams, and scenarios created for the student to practice hearing, reading, speaking, and writing these words.
Academic English I find is actually harder for me to predict students will struggle with and harder for students to learn. I cannot think of a lesson where the word “validate” will be a big focus for the day and I intentionally create materials or a mini-lesson to introduce it. Yet that word appears in the “Using Light to Study Planets” Article from my day 1 lesson plan. I am sure some of my students will not know it, but this might be the only time they see it in my class so I don’t think planning time to define it is a worthwhile use of class time. That’s why these words are harder, but students should also have access to tools (like dictionaries or translators) or strategies (like context clues or asking a partner) to help them negotiate meaning and move on.
As students’ conversational fluency increases they still need active support with literacy and specialized language
I find that typically around a WIDA composite score of 3 of 4 students begin to be able to speak fluently enough that they can sound to be proficient in English when their writing and reading skills are likely much weaker. I always check in with these students and make sure they have any help they need, because they are skilled enough to hide and pretend like they don’t need help to keep from standing out amongst their peers, even if they are struggling or stuck. I have a student right now who can hold a conversation really well right now, but their English vocabulary is very limited and their literacy skills are still developing. This means it is really easy to talk to them and get the impression they don’t need help, but they struggle to read and write a complete sentence so they still very much need scaffolding, different resources, or individual help in order to fully participate in most classwork.
Your students have pre-existing language skills, either in English or in their first language, use those skills to help communicate and support the growth of other skills
Most students learn to speak and listen more quickly than read and write, this can be useful. If I am working with my student from the last bullet point, who can speak well but not write well, then I can use that. For them it might be more appropriate to have a verbal test instead of a paper and pen one. Speaking to me will reveal how much they have learned in Earth Science class more effectively than them struggling to take a written test. Or if I am trying to support them when writing, I might ask them to speak with me until we’re both sure they know what they want to say and then ask them to write down what they said out loud to me. Last, your students can most likely speak (and maybe read and write) in their first language. It is okay to translate using their first language to communicate with each other while their English skills catch up. Newcomer students can really benefit from translator apps to speak to others, and translated materials to read and write complex thoughts while their English literacy is just starting. BUT be mindful that students should use both languages; their first language should not replace learning English, it should be a part of how they learn science AND English.
How to effectively use a student’s first language depends on their English proficiency and their L1 proficiency. Translanguaging is an approach to incorporate a student’s L1 into the classroom to both support their English development, but also to use and potentially strengthen their L1 proficiency. I like the way that Williams describes translanguaging in “Pedagogical Translanguaging”, “He (Williams) considers that translanguaging can use the stronger language to develop the weaker one so as to contribute towards a balanced development of the two languages”[2]. At times he criticizes how translanguaging is used in the U.S. merely to strengthen English development and not equally focused on L1 development. I think he’s right to criticize the US in that regard, but I also am not competent enough in any other languages to support my students’ growth in their L1. That said, translanguaging is a whole suite of educational tools and methods to help students use all their language skills to support their overall growth, and I recommend learning some of the basic principles to help your students be more successful, even if like me you’re not really able to support their growth in all languages.
Reading long and/or complex texts takes ML students more time and energy, if not being outside what they can comprehend on their own
As multilingual students develop their literacy skills it is important to pick / create texts appropriate for them. When selecting texts for students to read, all students should read texts with the same main content ideas. However, in order to make a text more accessible it is important to keep in mind a couple of things that can be changed to match a student’s skill level. Sometimes a text is just too long, and ML students take more time to read in English. So you can summarize a text and remove extra details. Oftentimes you will change length along with the complexity of the text. When changing the complexity keep in mind that you should change the complexity of the form; like removing compound sentences, metaphors, or switching to the active voice. However, you should also decrease the complexity or difficulty of the word choice. Removing enough academic English terms to bring the text closer to conversational English can make it more accessible. However, you should not remove all the complexity and all the challenge. It is about achieving the right balance and matching the challenge level to your students’ proficiency levels.
This idea of matching the challenge and complexity to a reader’s skill level is related to the Zone of Proximal Development (ZPD). This is the idea that you want to give students tasks that with the help of the teacher, their peers, or with tools and scaffolding the student is pushed to practice skills they are not yet able to do completely on their own. This prompts the students to push what they can do and learn. I really love this quote from Lantolf et al. as they describe the Zone of Proximal Development, “... the ZPD is forward looking through its assertion that the functioning one can reach in the present through cooperation with others is indicative of what one will be able to do independently in the future.”[3]
Sometimes what students need help with is putting their thoughts into words
Similar to the last idea, how it can be helpful to reduce the challenge of reading a text, sometimes what ML students need is a reduction in the difficulty of speaking or writing. Explaining how or why can be hard because they are very open questions, and while they do a good job showing what a student understands about a subject, they are quite challenging for a beginning learner to approach. Like you might do for a special education student, multilingual learners sometimes need narrowed choices in how they respond. Sentence starters provide the beginning phrases of a response and a student would write the latter half to finish the sentence with their own thoughts. If your student is really struggling with crafting a sentence though, it might be appropriate to give them fill in the blank sentence frames. These scaffold the writer even more by giving them most of a sentence which they fill in the key words to make it complete. It might be missing key adjectives or key nouns that the student is being asked to use to show what they know. A word bank can also be really helpful to scaffold a student’s writing because it removes some of the cognitive demand of remembering different terms but gives them the freedom to use the words as they please. The most narrowed responses might be something like matching or multiple choice, but at that point I would be nervous about the student’s reading skill being high enough to understand the prompt in the first place, so I’m not sure if that would be the most effective choice for truly beginner writers.
TESOL / ESL Teaching Approaches
In addition to the ideas above that I take into consideration when adapting lessons to meet multilingual students’ needs, I rely on two different lesson frameworks to help guide my lesson planning. One is called Sheltered Instruction Observation Protocol, or SIOP, which is more of a lesson evaluation protocol than it is a teaching approach. However, I use it like a checklist of what kinds of things should be present in a lesson and it helps guide my decision making. The other TESOL approach that I tend to use is called Task Based Language Teaching, or TBLT. TBLT is an approach designed around structuring authentic communication tasks that students need to use their language skills to complete. Let me briefly explain how I benefit from both ideologies below.
SIOP (Sheltered Instruction and Observation Protocol)
SIOP is an evaluation tool designed to rate lessons or observations based on if they have key factors shown to help multilingual students. These factors should in general be beneficial for all students, so it could be used to rate any lesson but it comes from TESOL research so it’s focus is meant to be for multilingual education. Like I said, I use it like a checklist or planning tool to help me keep in mind the kind of elements of a lesson that makes a lesson more accessible for kids and more powerful for helping struggling learners, both multilingual students, special education, and others alike. SIOP has 30 different items, but not all of them are actions you take during a lesson, so it’s more manageable than 30 sounds. That said, I feel like trying to fit all the 30 different items into a lesson is unproductive. Fitting all the items into every lesson really limits the creativity and flexibility of lesson design. While some of the items I try include in every lesson, others I think should be rotated throughout a week. For instance, SIOP encourages multiple types of groupings and interactions between students each day, which can force you to exclude more individual lesson days. Likewise not every lesson I plan includes hands on manipulatives, but I do think they are a good idea to include often. As a science teacher I do enjoy activities and labs, but not everyday can be hands-on. However, most of the items show up anytime you do good teaching, so it’s a useful guide for crafting a good lesson overall. My textbook from one of my master's classes did a good job explaining the items, how to achieve them, and why they are important. I would recommend it; “Making Content Comprehensible for Multilingual Learners, The SIOP Model” by Jana Eschevarria and others.[4]
Task-Based Language Teaching (TBLT)
TBLT is a language teaching approach that I borrow heavily from when refining a lesson. TBLT revolves around “tasks” which are communication activities that are designed to have students use content language in an authentic and required way. I first learned about Task-based Language Teaching from David Nunan’s “Task-Based Teaching and Learning”, where he defined tasks into two categories with, “I draw a basic distinction between pedagogical tasks, which are things that learners do in the classroom to acquire language, and real-world or target tasks, which are the uses to which individuals put language to do things in the world outside the classroom.”[5] I think a good task would marry the two categories. A task might be writing a lab report about the lab from yesterday; completing the task would definitely require using their literacy skills as well as their chemistry knowledge and vocabulary. A talk task however, might include a partner activity where students are deliberately given only part of the content they need to succeed and thus forced to talk to each other in order to cooperate and succeed. These are the kinds of tasks that I will admit I won’t do. I have a criticism of tasks that are arbitrarily difficult, and if completing the task becomes much more efficient if you break some arbitrary rules then the task begins to feel like a waste of your students’ time. I totally agree that structuring parts of a lesson to ask students to use their language skills to communicate in either speech or writing is good for your students. Talking about science ideas with each other is a good way for students to practice their language skills AND learn the science content; Improving both language and content skills is my goal after all. Talking, writing, and sharing are great ways for students to refine their ideas, learn from each other, practice their vocabulary, and improve their communication skills. So at my heart I try to implement the key ideas of TBLT into all my lessons, I just disagree with the definition of a “good task” outlined in TBLT sometimes. In fact, one of my major goals when revising the lessons for my unit plan was to create more structured opportunities for students to work together and talk!
Rationale for the Adaptations I made to Lessons
While I was planning the lesson plans for the unit, I had two overarching goals that should become clear as you read through my rationale per lesson. First, I wanted the students to talk more: to a partner, in a small group, or sharing across the whole class. So before I even started, I knew I wanted to create more structured opportunities for students to talk about astronomy. Second, last year most of the assignments in class did not flow well or were poorly scaffolded. So this year I wanted to change not just the handouts but whole lesson structures to make it easier for students to understand the why and the how of what we were doing in class. That said, the last theme you will notice in my rationale below is that I have fewer “ELL” marked students than before and my student’s language proficiencies are all relatively high. As I made changes to my lessons I made changes with my current students in mind, so you won’t see heavily accommodated versions of assignments, but I should explain below what kinds of changes I would make to support multilingual students with less English proficiency who need more support.
Day 1 - Spectra Lesson
The original version of the day 1 lesson was a very standard reading assignment. Students read the article, answered the summary questions at the end, and then as a class volunteers would share their answers for the whole class. Looking at it this year, I didn’t like it. At best a student would talk during the bellwork and then once more at the end if they volunteered to share an answer. So the lesson lacked communication, and it was solely focused around literacy skills. Not my ideal lesson for my students, multilingual or otherwise.
So I decided the reading lesson was too individual, unscaffolded, and did not focus enough on the types of spectra. I wanted students to talk and negotiate meaning more, especially about the different types of spectra which are a big take away goal. By switching the individual reading to a popcorn reading where we read together but students volunteer to read, students have the reader and me to help model the words and explain the meaning. Since we were now reading as a class we would stop and negotiate the meaning of each paragraph, which I liked because it let me model some critical thinking and active reading strategies when I narrated. Additionally, the popcorn reading meant that we finished much quicker. Students now had time to work in pairs to define and describe the different types of spectra, and then we had time for the short Identifying Elements Based on Spectra activity. And if you’ve read my philosophy page you will know that I want my students to do science, not just read about it, so I was really happy that students got to see and work with the spectra first hand.
I’ve been thinking though about how I would change the lesson or materials to support a low English proficiency multilingual student? Honestly, I am torn. The problem comes from working with students of a variety of skill levels in science and English. I think the bellwork, popcorn reading, pair sense-making, and then spectra activity is a great progression. However, the complexity and vocabulary of the reading feels like it would be above the skill level for students of WIDA scores 1 or 2, so simplifying the text seems like a good idea but then it would be unchallenging for most students. I really liked the group popcorn reading because it allows us as a class to negotiate meaning, talk about the text, and model literacy skills. However, that really only works well if we are all reading the same text. Unless I had a helper to lead a second reading group with a different proficiency level text, I think keeping the text unsimplified with the knowledge that my lowest proficiency readers won’t understand all the text is still the optimum choice since it allows us as a whole class to read together and do all the literacy and cognitive skills I wanted to model with the students. That said, I stand by the rest of the lesson and think this version is much better.
Day 2 - Star Spectra Lesson
This lesson is adapted from a recent Earth Science lesson. I adapted the language of the instructions to be simpler and more scaffolded, my native speakers were struggling so I wanted to scaffold this activity more for my ML students especially. It is important that multilingual students can understand instructions enough to know what to do and how. Like the Day 1 lesson, I also changed the activity to encourage more student interactions. This time I turned the whole thing into a partner exercise that resembles task-based language teaching. Students now work in pairs and share the job of gathering data and classifying the stars, and then talk to each other to fill in each other’s data tables. In the second half the difficulty ramps up, so I deliberately have them work in pairs to read through the descriptions and decide which one matches their star. Again, each pair makes a whiteboard and shares out what they decided to the whole class. So I am using grouping to both help scaffold more difficult tasks by having team mates, but also using groups to create opportunities for them to communicate with their partner and the rest of the class about what they are learning.I’m really proud of how by changing the grouping of the assignment it helps scaffold more difficult parts but also creates the student interaction that I look for.
If I wanted to match the lesson to a WIDA 1-3 level, I think the tasks they carry out are still good. However, I think modelling the steps of what to do would be even more crucial, and I would definitely leave an example on the board. I do worry about the latter half though when students are assigned an unusual star to describe. There is more academic vocabulary in this section and many uncommon but key words for understanding the star descriptions. I wonder if instead, it might be smart to give students a star, ask them to describe how it’s spectrum is unique, and then ask them to decode their star’s description to see how well their findings match. It’s hard to impossible to simply describe what these stars do and why, so instead of simplifying the text I think the optimum route is to reduce the length of text students read (like I suggested up above in my section about “reading long and/or complex texts”).
Day 3 - Black Body Radiation
This assignment I was pretty happy with originally. After an intentionally crafted bellwork to connect to prior learning, I gave a brief introduction into the concept for the day. Then I modeled how to read the black body radiation graph and how to calculate the temperature of objects. Students completed the assignment successfully (mostly as small groups) while I helped out with the demonstrations in the back. Splitting students up to do the demonstration allowed them to experience core ideas first hand, but also allowed them to ask questions about the demonstration, black body radiation, or the assignment. The biggest change I made was how we went over the assignment at the end. This time I split the class into five groups, which were each responsible for a different section of the assignment. I split the class so that no one was whiteboarding with students from their table group, so they had to mix and come to a consensus with people from other groups that might have different responses from them (this is inspired by one of the tasks from TBLT). Additionally, I was careful to give the simpler prompts to my groups with people who struggle more and my more complex open ended prompts to the groups with more advanced students. So that whiteboard groups were discussing and sharing prompts that matched their language skill levels.
However, if I did this again, I’ve already decided that I want more scaffolding to the beginning introduction part. While my students did better than expected during this assignment, I think they would have benefitted from a brief guided notes section at the start. If I had students with lower language skills in the room that would have been important for them to initially learn about black body radiation and record notes to act as a guide and reminder later in the lesson. I think a roughly half page where we label a black body radiation spectrum together with fill in the blank sentences to record key rules and trends about these spectra would be more ideal. So I’ve decided to change how I teach Black Body Radiation next year.
Day 4 - Redshift / Blueshift
I did not like the original version of my redshift lesson. It was just a bunch of number crunching that I didn’t feel really encapsulated all the little things that redshift / blueshift plays a role in astronomy. Nor did the lesson encourage students to interact and practice their language skills other than their graphing literacy. So I wanted to make it a handful of smaller stations to show students the different facets of redshift / blueshift.
I added the slides at the beginning to introduce the concept of redshift and blueshift formally and to increase comprehensible input as I did so. I then split the one large original task into two stations and found two more ideas about redshift to create two more stations. I incorporated something about binary stars and exoplanets because students were really curious about them at the end of the day 2 lesson. For each of the stations I made the instructions explicit and step by step to make them easier to read and follow, especially for my students who struggle with larger tasks. I tried to find helpful images for each station, but I think the distance station could still use one. The images are there to act as another mode for readers to understand what the text is telling them; essentially the images are for increased comprehensible input. In the instructions, students are tasked with working in small groups to solve each station, however the task design does not require multiple students to actually work together, which could be changed for next year to increase student interaction. The majority of my students did work in small groups and benefitted from it. Again, I was surprised at how well my students did; I thought they would struggle with the math more given what I saw earlier in the year. So while I think the new lesson is improved, I would make more changes for next year.
Next year I want to find more images to add to the stations; maybe a video. These would especially help with lower English proficiency students. I also want to add one more station about binary stars which I did not find good data to use, but I know it is out there. Kids were curious about binary stars because of the Day 2 star spectra they saw and I want to help them explore that curiosity. If I had lower proficiency MLs they would definitely need more support. A print out of the introductory slides to use as a reference during the stations might be helpful. However, the text in the radial velocity article and the historic galaxy data stations should be simplified for new ML students. While reviewing them before class, I made the intentional choice to not change that text this time because it felt like an appropriate amount of challenge and length for where my kids are at right now. However for a beginner English speaker the radial velocity article would be too long, and its vocab probably too academic, so that article I would simplify for them. The historic galaxy data station includes text that is only a paragraph or two, but the text is very formal so the grammar and word choice are a challenge for most of my students to understand at first. The galaxy data station I would definitely re-write to be more student centric language for new ML students in order to bring the text closer to their proficiency level. If I also rewrote the stations to focus on student groups, then MLs would also get support from the higher English proficiency group mates during next year’s stations activity.
Day 5 - Scanning the Spectrum
While constructing the unit, I realized that students did not write very much. I had done a good job of achieving my goal to increase student speaking interactions, but students were hardly asked to write. So I decided to finish the day 5 lesson with a large (3 paragraph) writing task to practice their literacy skills.
I adapted this lesson from near the end of the semester and brought it much earlier. The original version assumes students know a lot about galaxies, so I broke the overall assignment into multiple small steps with very explicit goals to make it more approachable. Each small step is “narrower” than before and structured to guide the student as they look at different images and describe what they are looking at. Where before the prompt might have said something like, “What does this image tell you about the galaxy?” Now the prompts are broken down to something like, “Where in the galaxy is this light coming from? What types of objects produce this light? Where are these objects in the galaxy?” There are tables to help students negotiate what settings to use, as well as what types of objects go with each wavelength. All the pieces they will need at the culmination of the assignment I either give them or explicitly ask them to find. The small steps for data collection are themselves a scaffold / guide for writing the summary paragraphs at the end of the assignment. I am really proud of how I structured the data collection stage because it acts as a checklist of what to include in their paragraphs. I deliberately point that out to students during the lesson to help them know what to include and in roughly what order when they begin writing.
For newer English speakers, an altered version of the assignment with sentence starters or sentence frames to guide their data collection would be appropriate to help students understand what they are looking at. Likewise structure to help students write their paragraphs is a good choice to support lower proficiencies. It also might be appropriate to ask students to write fewer paragraphs. The exact level of scaffolding when writing depends on the student’s proficiency level. To help support students when data collecting a set of sentence frames like:
- The object that glows in this type of light is ______________
- The ______________ of the galaxy is making this type of light
- Therefore ____________ must be in ________________ of the galaxy
And to support students when writing their summary paragraphs, a higher proficiency ML could use the same prompts as what I have now, but a middle proficiency student might benefit from a checklist like:
Make sure to include:
what Visible light shows in the young galaxy
what Ultraviolet light shows in the young galaxy
what Infrared light shows in the young galaxy
what X-ray light shows in the young galaxy
what Visible light shows in the old galaxy
what Ultraviolet light shows in the old galaxy
what Infrared light shows in the old galaxy
what X-ray light shows in the old galaxy
How the young and old galaxies are shaped differently
How the young and old galaxies have different Visible light objects
How the young and old galaxies have different Ultraviolet light objects
How the young and old galaxies have different Infrared light objects
How the young and old galaxies have different X-ray light objects
However to support a really low proficiency student it might be appropriate to lower the complexity down to fill in the blank kind of statements. And at this point I would be tempted to drop the compare and contrast section too.
In the young galaxy,
the visible light show ____________________ in the galaxy’s _____________
the ultraviolet light show ____________________ in the galaxy’s _____________
the infrared light show ____________________ in the galaxy’s _____________
the X-ray light show ____________________ in the galaxy’s _____________
In the old galaxy,
the visible light show ____________________ in the galaxy’s _____________
the ultraviolet light show ____________________ in the galaxy’s _____________
the infrared light show ____________________ in the galaxy’s _____________
the X-ray light show ____________________ in the galaxy’s _____________
References
- Gass, S. M. (2015). Chapter 11. Comprehensible input and output in classroom interaction. In N. Markee (Ed.), The handbook of classroom discourse and interaction, (pp. 182–197). John Wiley & Sons, Incorporated.
- Cenoz, & Gorter, D. (2021). Pedagogical Translanguaging. Chapter 3. DOI: 10.1017/9781009029384
- Lantolf, J. P., Pohner, M. E., Thorne, S. L. (2020). Chapter 10. Sociocultural theory and L2 development. In B. VanPatten & J. Williams’ (Eds.), Theories in second language acquisition: An introduction (3rd ed.). Routledge.
- Echevarría, J., Vogt, M., Short, D., & Toppel, K. (2023). Making Content Comprehensible for Multilingual Learners: The SIOP Model (sixth edition). Pearson.
- Celce-Murcia, Brinton, D., & Snow, M. A. (2014). Teaching English as a second or foreign language (Fourth edition): Nunan: Chapter 29 pp.455 - 470