A Course in English Language Teaching (2012) - Teaching Writing - PDF

11 Teaching writing It is asynchronous, or time-independent. We usually read text some time after it has been written. Even with ‘synchronous chat’, there is a time-lapse between production and reception. Spoken discourse, in contrast, is in most cases produced and recei...

11 Teaching writing It is asynchronous, or time-independent. We usually read text some time after it has been written. Even with ‘synchronous chat’, there is a time-lapse between production and reception. Spoken discourse, in contrast, is in most cases produced and received simultaneously. The person or people being addressed are not physically present. The target audience for a written text – whether a single addressee, closed group or the public at large – is rarely physically present. Spoken interaction is mostly face to face. It is produced slowly. Writing is much slower than speaking, reading or listening. This is one reason why it is used least of the four skills in real-life communicative activity. It is a learnt and high-prestige form. Most people acquire the spoken language (at least of their own mother tongue) intuitively, whereas the written form is normally taught and learned in school. Literate people are more respected, in general, than the illiterate. It uses more standard forms. English speech typically varies widely, in accent, lexis and grammar, according to the cultural or linguistic background of the speaker; writing, in contrast, is more uniform and tends to observe conventional usages (formal or informal). Some implications for teaching We have seen that writing cannot normally be ‘picked up’ but has to be systematically taught. This means that we actually need to devote a lot of attention to teaching it, even though it is actually used by most people far less than the other skills. In speech, students express their ideas in linear fashion, as they occur to them; in writing, they have time to rewrite and edit. So we need to help them ensure, through such rewriting and editing, that their writing uses acceptable language and that the text is organized and coherent (see Section 11.5 below). 11.2 Formal and informal writing Most writing is formal. Stories, reports, most webpages (wiki entries, for example), newspaper articles, fiction, the book you are reading at this moment … all these are formal texts. Informal writing was in the past only used for quick notes or reminders; but these days it is used much more: in online ‘chat’ and texting (SMS), for example. Here is an example of informal writing: in this case a ‘chat’ between friends about an upcoming wedding (between Jane and Joe): Emma says: taxi? to register office? yes? no? Emma says: unless you’d rather walk Jane says: if raining, then yeah, a taxi A Course in English Language Teaching 151 11 Teaching writing Emma says: excellent Emma says: anything you need for the day that we could bring? Joe says: me Jane says: nooo (Crystal, 2006, p. 251) Task List the differences you can find between formal writing, using the example of the text of this book, and informal writing, using the example above. Then read on. Some differences between formal and informal writing are listed in the table below. Formal writing Informal writing Grammar and lexis The text follows standard The text may not use full grammatical rules and sentences or formally ‘correct’ uses conventional, formal grammar and spelling; uses vocabulary. colloquial vocabulary (e.g. yeah for ‘yes’) and contracted or abbreviated forms like you’d. Punctuation Punctuation follows standard On the whole, punctuation rules. is used less. Some ‘special effects’ punctuation may sometimes be inserted, for example rows of exclamation marks for emphasis. Detachment The writer is detached from The text is usually targeted at the reader personally, in a specific audience, personally time and in space. The text known to the writer, often read is therefore not normally immediately, or a very short designed to be immediately time, after it is written, and responded to. responded to quickly. Editing and The text is written first as The text is not normally edited redrafting a draft, and then edited but sent to the recipient and redrafted to polish as a ‘first draft’: it may the language, style and therefore often include slips organization. or ‘ungrammatical’ usages and will not have a systematic structure. 152 A Course in English Language Teaching  11 Teaching writing Formal writing Informal writing Length Most texts are more than a Most texts are short: a paragraph in length: reports, few words, a sentence, letters, articles, stories, even occasionally more. complete books. Explicitness The text is explicit: little The text may be largely shared knowledge is taken implicit, assuming a for granted and all necessary large measure of shared information is provided. knowledge; there will therefore be ellipses (missing words assumed to be easily guessed by the reader), and pronoun or other references to real- world subjects may not be explained. Formal and informal writing are presented in the table above as two distinct types or styles: but of course, there are gradations: some emails and blogs, for example, are written semi-informally, with characteristics from both columns. In teaching, however, it is important to make students aware of the difference in principle, and in what contexts and circumstances each may be appropriate. They should know, for example, that they should not use informal vocabulary or contracted forms like can’t or u (for ‘you’) in formal academic writing. Most writing is formal, and students need to learn the necessary skills and knowledge to enable them to compose formal texts of different kinds. If they are able to do so, we will not have much difficulty teaching them the specific ‘shortcuts’ and stylistic peculiarities of informal writing – but the converse is not true. So it is probably best in teaching writing to focus mainly on the formal mode. 11.3 Beginning writing: the letters Note. This section is appropriate for monolingual classes whose L1 uses a non- Latin alphabet or another writing system. Such classes are very often composed of young learners (aged between 7 and 11). Task Think back to when you yourself were taught to write Latin script – whether in your L1 or when you learnt English or another Latin-alphabet-based language. How were you taught? Can you remember what was useful (or not) in helping you to learn letter formation? Some basic aspects of the teaching of writing to beginners apply equally to the teaching of reading, and have already been dealt with in Unit 10: Teaching reading, pp. 136–42. As with reading, it is important to know some simple A Course in English Language Teaching 153 11 Teaching writing conversational English before beginning to learn the letters; and we are likely to teach single letters before we teach letter combinations in words. We will probably teach the most common and useful letters before the less common and useful ones, and cover the lower- and upper-case forms at the same time. But there are other skills students need to master that are specific to handwriting and, later, typing. Here are some of them, with practical teaching implications that you may have already thought of if you did the task above. Direction Other writing systems often go in a different direction from English: from right to left instead of from left to right (Arabic, Hebrew) or vertically (Chinese). This involves not only getting used to moving one’s hand in a different direction along the line, but also often learning to form the letters in a different direction. For example, speakers of Arabic or Hebrew are used to drawing circular letters in a clockwise direction and will now have to learn to form them anti-clockwise. If they are not deliberately taught otherwise, they will continue to write these letters clockwise, which will slow down the flow of handwriting and make it difficult to join up letters should they wish to do so later (see Cursive writing below). Practical implications We need to provide students with models of correct-direction writing: by modelling the letter writing on the board, and perhaps also by providing the alphabet written out with little arrows showing in what direction it should be written. It can help as a preliminary exercise to get the students to write rows of waves or loops, running from left to right, as follows: Then they need to practise writing out rows of similar letters, and later combina- tions of different ones in words, while the teacher makes sure they are forming them correctly. 154 A Course in English Language Teaching  11 Teaching writing Height, depth and level One thing students learning the Latin alphabet often find tricky is getting the height and depth of letters right: making sure that letters like d, l, b in fact have ‘arms’ that are of a similar height to capital letters, and that letters like p, y, g have ‘tails’ that are sufficiently long beneath the line. Some students also have problems with maintaining level horizontal writing. In both cases, they need the guidance of ruled lines at the early stages of writing. Practical implications Early writing should be done within parallel lines, which limit the height of letters and make sure they are level. These have two parallel lines in the middle which limit the height of the smaller letters such as c, m, o, and two added lines above and below to show how far the taller and deeper letters should reach. Cursive writing There is some debate over whether or not to teach students to write cursive (joined-up) script. If their L1 uses the Latin alphabet, then this will depend how they have already been taught to write it. If, however, the Latin alphabet is a new writing system for them, you will need to decide whether or not to teach cursive. Sometimes the decision will be taken for you: there is a clear general policy in your school, and perhaps in the country in general. Elsewhere, it is more a question of clarity (non-cursive) versus more attractive appearance (cursive). Practical implications The overriding criterion here has to be comprehensibility; and letters written separately are more likely to be legible than joined-up letters. A possible compromise I used in my own primary classes was to insist that everyone used non-cursive for the first year. In the second year I taught them cursive, and anyone who wished was told that they could continue to use it provided it was clearly legible. Most of my students in fact reverted to the non-cursive and later, as adults, used a combination, joining up some letters and not others. A Course in English Language Teaching 155 11 Teaching writing Typing For the foreseeable future it will still be necessary for our students to know how to handwrite in English: for personal greetings, form-filling, quick notes, shopping lists, etc. But it is becoming more and more necessary for them to learn to type in English as well, as most writing today is done on a computer. In my opinion, they should be taught to touch-type in their own language in elementary school: but for some reason, even where computers are available, this is rarely done. Practical implications Normally there is no time in an English course to provide the hours of practice needed to acquire touch-typing skills, though we might suggest that they teach themselves using a computer or internet-based program (you can find free programs online by searching for ‘touch type’). In any case, we should give our students plenty of opportunities to use the keyboard, as well as a pencil or pen, to do writing exercises, in order to increase their typing speed. If they have already learnt this elsewhere, it is also useful to teach them how to use word processing tools to format their compositions: different fonts, sizes and spacings, different colours and positions. Basic copying or simple composition exercises can be made more motivating by allowing students to use these tools to improve the presentation of a text. Speed and legibility When teaching writing, the two main aims are to enable students to write reasonably fast, and to write legibly. However, emphasis on the one may be at the expense of the other. If you write very fast, your writing may be difficult to understand (true of typing as well as handwriting!). If you write very carefully, so as to be legible, you may sacrifice speed. Practical implications Having taught the letters, we need to give students plenty of practice in using them to write words and sentences so that they get to do so faster, but not at the cost of legibility. See Beginner writing tasks below for some ideas. Later, any grammar or vocabulary written exercises will obviously give practice in speed and accuracy of handwriting or typing as a useful side benefit. Beginner writing tasks Apart from the first two ideas, which clearly relate to handwriting, all the following task-types can be done either on paper or on the computer. Copying. Copying is a useful way for students to practise letter formation with the ‘scaffolding’ of a model. It can be done by tracing (using tracing paper or following dotted lines) or copying lines of letters or words. Colour copying. Since early letter-writing practice can be boring, invite students to use different-coloured pens, or decorate their exercises with coloured frames or underlining. Thoughtful copying. Another way of making early writing practice interesting is to ask students to copy according to particular criteria: to copy, for example, 156 A Course in English Language Teaching  11 Teaching writing only words that are names of animals (or any other lexical set you choose); or in a different order; or into different categories; or in order to label pictures. Transliteration. Students transliterate single letters or words (cognates or names of people or places are particularly useful) from their L1. Dictation. Students write down single letters or simple words from dictation. Completion. Students fill in the missing letter(s) from a known word, perhaps illustrated by a picture. Labelling. Students label pictures with simple phonetically spelt words. 11.4 Tasks that promote fluent writing Most of students’ writing in an English course is not done primarily in order to develop writing skills, but because writing is a convenient means of engaging with aspects of language. For example, students write down new vocabulary; copy out grammar rules; write out answers to reading or listening comprehension questions; do written tests. In these examples, writing is used as a means of getting the students to attend to and/or practise a particular language point, to make a note of new language for later reference, or as a convenient method of testing it. Fluent writing tasks, in contrast, aim to improve students’ ability to compose written text for communicative purposes. Aspects of accuracy (grammar, vocabulary, spelling) are, of course, important in formal writing, as discussed earlier in this unit; but the main focus is on meaningful writing following the conventions of a particular genre. (Tasks that focus on the forms rather than the communicative purposes of writing will be discussed in Section 11.6 below.) Some criteria for the planning or selection of writing tasks are: Interest. The task should be motivating and stimulating. Level. The language required should be appropriate to the level of the class. Relevance. At least some of the tasks should be similar to the kinds of things students may need to write themselves, now or in the future. Simplicity. The task should be easy to explain. Often the provision of a model text can help to clarify. Personal appropriateness. The task should be one that you, the teacher, feel comfortable with and that fits your own teaching style, goals and preferences. Task Choose four or five specific writing tasks from the seven categories below, and think about how they might work with an intermediate-level class (if possible, one you are familiar with yourself). Suggest criticisms and any improvements or additions you can think of. Then read on. A Course in English Language Teaching 157 11 Teaching writing Writing tasks 1. Creative writing a) A story based on some kind of given stimulus: a title, a picture or series of pictures, or a first or last sentence. b) A personal anecdote describing an occasion when you were disappointed (or afraid, surprised, relieved, etc.). c) A poem based on a given stimulus: topic, a particular structure, first or last lines. 2. Instructions a) An instruction sheet for something you know how to do (for example, prepare some kind of food). b) Directions how to get somewhere. 3. Interpersonal communication a) A letter or email applying for a job. b) A letter of complaint. c) A reply to a letter. d) A comment on a blog: either one that already exists, or one set up for the class. e) An email telling a friend what you’ve been doing recently and suggesting a meeting. 4. Description a) A description of a view, a place or a person. b) A description of a process, such as a scientific experiment, the life-cycle of an animal, a sequence of events represented by a flowchart. 5. Responses to literature a) A synopsis of a book, play or film. b) A review evaluating a piece of literature the class has read. 6. Persuasion a) A recommendation for some kind of change in your home community or place of work/study, addressed to the appropriate authority. b) An advertisement for a product. c) A leaflet promoting an institution, a tourist attraction, a course. 7. Information a) A newspaper report on an item of news, genuine or imaginary. b) A short paper providing information on a particular person, event or invention, previously researched on the Internet. Comment 1. Creative writing. Some students respond well to tasks that demand imagination and creativity: others really don’t like them! For the less creative students, you will need to provide much more structure and support. Poems are surprisingly easy and pleasurable to write if based on an appropriate stimulus: see some excellent practical ideas in Writing Simple Poems (Holmes & Moulton, 2001). The results of such tasks are often enjoyable for other students to read: post them on the class website, or on a noticeboard in the classroom, or read them out to the class (with the author’s permission, of course). 158 A Course in English Language Teaching  11 Teaching writing 2. Instructions. These tasks may be interesting for students if they relate to processes or places they know a lot about. They may require some preliminary vocabulary review or teaching; and you may wish to give some advice on the layout of instructions: numbered steps, for example, or illustrations where necessary. They are particularly useful for classes in English for Specific Purposes, such as engineering or nursing. 3. Interpersonal communication. Assignments like a letter or email applying for a job or a letter of complaint are probably most suitable for adults or older teenagers. Students also need to be aware that such texts would demand far more formal English than the blog comment or email. If they are written as paper letters rather than email, then you will need to teach some conventions of letter writing: address, date, addressee, formal beginnings and endings. The blog task is often highly motivating, and can continue later, with other students adding further comments. 4. Description. This can be done at a fairly basic level of proficiency. To make it even easier, you might ask for phrases or single words, rather than full sentences in a coherent paragraph. For the personal description, it is helpful to provide in advance some topics that the writer might relate to: appearance; occupation; personality; interests; life story. Apart from the flowchart, descriptions of processes can often be laid out in other ways, such as tables or graphs. 5. Responses to literature. Just providing a synopsis can be rather boring. The review is more difficult, but more interesting to do, with the challenge to analyse and criticise, and the purpose of encouraging (or discouraging) potential readers or viewers. 6. Persuasion. The recommendation task is suitable for rather more advanced classes as it demands fairly careful planning of content and organization. The advertisement may be easier and can be decorated with coloured fonts and designs, and illustrated by pictures. The leaflet may be done collaboratively, each student in a group writing a different section: and again there is the possibility of illustration and decoration. 7. Information. The newspaper report can be based on a model: an authentic news report which has been read in class. The short paper is a first step on the way to academic writing, though it may be done at intermediate level. It can be highly motivating if the student is researching something that interests him or her personally. Note that you may need to do some preliminary instruction not only on the need for formal language, but also on the need for structure: introduction, headed sections, conclusion. A Course in English Language Teaching 159 11 Teaching writing An alternative is to require such assignments to be formatted as presentations with slides. In this case, there is less actual writing, but the student will need to be careful with the choice of headings and notes to be shown on the slides, and with the formatting: size of font, line spacing, use of punctuation and so on. Writing in class After having explained the task, you are likely to send the students to do most of the writing on their own at home. However, it is useful to be able to vary this routine by occasionally doing shorter writing assignments in class. For example: Collaborative planning. Students plan their texts in pairs or small groups in class, before going off to do the actual writing alone. They later come together to compare results. Five-minute writing. Students have exactly five minutes – no more, no less – to write something in class. The time limitation usually increases concentration and is a useful quick way of providing extra writing experience. Multiple contributions. Students are given a topic and are asked to write a phrase, a word or a sentence on it at the top of a sheet of paper. They then pass the paper to their neighbour, who adds another line, continuing the idea which the first student has expressed (not just adding another sentence about the topic). The process continues until there are 5 to 15 contributions. Note that the paper remains open, not folded, so that every new writer can see all the previous contributions. The results are then read out to the class. The same can be done with a story: the first line is given, and students continue adding further events to the narrative. Practical tips 1. Length. Give students a clear idea of how long you intend the composition to be. If they are writing on computers, then they can easily check the number of words; if not, an A4 page of double-spaced lines is about 300 words. As a rough guide, beginner classes can be asked to write about 50 words, intermediate classes between 100 and 200, and advanced classes 300 or more. 2. Rewriting. Get students to rewrite their compositions at least once: see Section 11.5 below. 3. Collaborative writing. Most writing is best done by students working on their own; it can be planned collaboratively, but if you ask students to do the actual writing together, the result will usually be that one of them does all the work and the other(s) is (are) passive! For collaborative editing using wikis, see Section 11.5 below. 11.5 Process writing When they have mastered the basics, students need to progress and improve their writing. This can be helped to some extent by focused instruction on spelling and punctuation (see Section 11.6 below); however, the best way to promote students’ writing ability is through the process-writing cycle: the experience of writing a first draft, getting feedback and rewriting. The main problems for teachers are how 160 A Course in English Language Teaching  11 Teaching writing to help students write their first draft(s); and how to give effective feedback on these that will help them produce as good a final draft as possible. F i n a l d r af t Edit First draft No assessment Fe e d b a ck Students need to be aware that the first draft is not Second draft graded and nor are any other preliminary drafts (if the process-writing cycle is repeated more than once). The Edit assessment is given only on the final draft. Feedback Not giving a grade for preliminary drafts has two First draft important results. First, it lowers stress: students feel freer to experiment and to use language they are not quite sure of but want to try out: they know that they will not be penalized if they get it wrong and will be shown how to get it right. Second, they are highly motivated to implement feedback and improve in order to achieve a better final draft. (For some detailed guidance on later assessment of the finished assignment, see Unit 12: Assessment and testing, pp. 181–2.) Support Having given the assignment, we need to provide support in order to ensure that students write their first draft as well as possible. This support can include: providing key vocabulary that you think students might need or that they ask for; providing a ‘model’ text similar to the one required by the task; some discussion of possible content; guidance on the organization of texts of the relevant genre, and occasionally conventional formats (e.g. the placing of the address or date on a paper letter); allowing the beginning of the writing in class so that students have the opportunity to consult you as they write. They can then continue at home. Feedback FAQs Below are some Frequently Asked Questions (FAQs) that relate to feedback on process writing, an analysis of the basic problem in each case, and some advice. 1. What should feedback be mainly on: language? content? organization? The problem. When a student submits a piece of original writing, the most important thing about it is, arguably, its message: does it succeed in conveying the content required in the task? Then there is the organization and presentation: are the ideas arranged in a way that is easy to follow and interesting to read? Finally, there is the question of language forms: is the grammar, vocabulary, spelling and punctuation of an acceptable standard of accuracy? Many teachers are aware that content and organization are important but find themselves focusing mainly on language forms in their feedback, conveying A Course in English Language Teaching 161 11 Teaching writing the implicit message that these are what matter, sometimes to the exclusion of the other aspects. There are various reasons for this: 1. Mistakes in spelling or grammar catch the eye and seem to demand to be corrected; they are very difficult to ignore. 2. Students also usually want their language mistakes to be corrected (see Unit 7: Error correction, p. 93). 3. Language mistakes are far more easily and quickly diagnosed and corrected than problems of content and organization. Advice. We should correct language mistakes. Our problem is how to do so without suggesting that this is our only (or main) basis for evaluation, and the only aspect which needs to be improved in a second draft. So corrections to language or style should be noted, but we should also take care to include comments on content and organization. 2. Should all mistakes be corrected? The problem. If we accept that language (including punctuation) should be corrected, another problem arises: should all language mistakes be noted, even if there are so many that the page will be covered with corrections? If not, how do we judge which to relate to and which not? Advice. The problem is one of potential conflict between two of our functions as teachers: language instruction versus support and encouragement of learning. Correcting mistakes is part of the language instruction, but too much of it can be discouraging and demoralizing. Also, over-emphasis on language mistakes can distract both students’ and teachers’ attention from the equally important aspects of content and organization, as noted above. The answer is obviously some kind of compromise, which will vary according to the course objectives, class and student. We might correct only mistakes that actually affect meaning (i.e. might lead to misunderstanding or confusion), and/or those which are very basic. Or, of course, we can vary our response according to individual need. In any case, it is important to ask the students themselves (even younger ones!) how, and how much, they want to be corrected. 3. Should we let students correct or give feedback on each other’s written work? The problem. Correcting written work is very time-consuming, particularly with large classes. It helps to let students correct and edit each other’s writing. They may not be able to identify all its good or bad qualities, but they will detect at least some of them. The problem is: will students feel uncomfortable correcting, or being corrected by, their classmates? Will they accept criticism (positive or negative) from each other? Advice. Students on the whole, it appears, prefer to be corrected by the teacher rather than by their peers (see Unit 7: Error correction, p. 94). On the other hand, peer correction can be a time-saving and useful technique. Also, from the point of view of the ‘peer-corrector’, critical reading for style, content and language accuracy is a valuable exercise in itself. See Practical tip 4 below for the use of wikis. 162 A Course in English Language Teaching  11 Teaching writing Practical tips 1. Give feedback quickly. Students get much greater benefit from your corrective feedback if it is given immediately, or very soon, after they have submitted their work. They should also be required to rewrite and re-submit within a fairly short time limit. 2. Use ‘track changes’. Word processing on a computer means you can make changes or corrections and add comments to a document, while the original text can still be clearly seen. This is a very useful and time-saving way of correcting. Similar editing tools are available on PDF document readers as well. 3. Use ‘share documents’. If you want your students to do multiple rewrites, then instead of sending documents by email attachment, try using a file hosting service that allows you to share documents online (for example, Windows Live Skydrive, Dropbox or Google Docs). Your student uploads a document and names you as a ‘sharer’: you can then annotate or correct it, and the student can immediately see what you have done and implement the corrections in a second version, which again you can see immediately. LMS programs also allow you to do this (see Unit 16: Classroom interaction, p. 241). 4. Use wikis for collaborative editing. The use of the wiki for editing makes the whole process much easier, more collaborative and more learner-friendly. A text is uploaded onto a class wiki, and then any student, including the author, can insert changes or suggestions, which can then be immediately checked by the teacher. Previous drafts are preserved in a ‘history’ file within the wiki, so you can always go back and check an earlier version. 11.6 Spelling and punctuation Spelling Contrary to general belief, the large majority of words in English (about 84%, according to Pinker, 1995) are actually spelt either phonetically or according to regular rules which are fairly straightforward to teach and practise. This means that teaching students to spell correctly is not as difficult as you might have thought. And most of the words that have irregular spellings are very common ones which students learn anyway very early on: the, to, what, one. Other irregular spellings can be taught as the individual words come up. Some basic spelling rules that are worth teaching are: the digraphs th, ch (tch), sh, wh and the less common ph; the final e which causes a previous vowel to be pronounced like its name, as in late, these, time, hope, tune; the letter c, usually representing the sound /k/, regularly pronounced /s/ before i, e, y; the suffixes -tion, -sion, -ssion; the prefix al- spelt with one l in words like always; the suffixes al, ful spelt with one l; the letter g, usually representing the sound /g/, but sometimes (not always) pronounced /ʤ/ before e, i; the u following q; A Course in English Language Teaching 163 11 Teaching writing ck instead of c at the end of one-syllable words; the combination -all in short words, pronounced /ɔ:l/; the combinations -ight, -ought. Advice to students As well as teaching these rules, it is quite useful to give students advice on some generalizations that apply to a variety of words. The letter z is rare: usually when you hear the sound /z/ it will be spelt with an s. The letter j is rare: usually when you hear the sound /ʤ/ it will be spelt g or dg/dge. The letter k is relatively rare: usually the sound /k/ is represented by c (except at the end of short words, where it is likely to be ck, see rule above). A double consonant usually causes the previous vowel to be pronounced short, not like its name, in words like apple, filling (compare paper, filing). Hence the rule about doubling the consonant when adding the -ed, -ing suffix to short verbs, or making the comparative of short adjectives. Practice tasks for spelling Dictations. Dictate a set of words that you have taught which follow a rule the students have learnt. Other variations are as follows: Dictate a set of words that the students don’t know yet, but whose spelling follows a rule they know, and challenge them to spell them right. Provide the students with the target words, but with some key letters missing. You read out the full words, and they fill in the missing letters. Dictate only the definition of a word, students write down the word. Provide the L1 translation, students write down the English word (only, obviously, if this is a monolingual class whose L1 you know). For more variations on dictation, see Dictation: New Methods, New Possibilities (Davis & Rinvolucri, 1988). Recall and share. Write the target items on the board, give students a minute to look at them, and then delete them. The students try to remember all of them, first individually and then sharing. Finally, you display all the items again. This is particularly useful for words spelt irregularly. (It is also an appropriate exercise for vocabulary consolidation: see Unit 5: Teaching vocabulary, p. 48.) Task Add three or more spelling activities to those given above. They can be ones you experienced as a student yourself, or from a book, or of your own creation. Punctuation and capital letters The most common punctuation signs are likely to be used in the students’ L1 in a very similar way; for example, the full stop or period (.), the comma (,), the question mark (?) and the exclamation mark (!). There may, however, be marked differences in the way quotation marks (single or double, ‘ ’ or “ ”) are used. There are also specific punctuation usages in some other languages which are different from English. Spanish adds an upside-down question mark at the beginning of 164 A Course in English Language Teaching  11 Teaching writing questions, for example, and German inserts a comma before the equivalent of ‘that’ in relative or noun clauses. If you are teaching a monolingual class whose language you know, you will probably be aware of such differences and will teach them as they occur in reading texts, or as errors in students’ compositions. Students whose L1 does not use the Latin script may have problems mastering the use of capital letters to mark the beginning of sentences and proper nouns: these will need some focused teaching at the early stages. See the Practice tasks below for some ideas. Task Think of another language you know. What similarities and differences in punctuation have you noticed between English and that language? What difficulties might a native speaker of this language have in learning English punctuation? Practice tasks for pronunciation and capital letters Inserting punctuation. Give students a text with no punctuation or capital letters and invite them to correct it. It is, perhaps, best with most classes to provide texts from which only specific items have been excluded. For example, there are no capital letters, or no commas; or no quotation marks in a conversation. Students are told which type of item is missing and to insert them where appropriate. Capitals. Dictate a mixed list of common and proper nouns. Students write them down, inserting the initial capital where appropriate. Dictation. Dictate a short and fairly simple text, where the spelling is not a problem but which needs quite a lot of punctuation and capital letters. Recall and share. As in the ‘recall and share’ activity suggested for spelling, give the students two minutes to look at a text similar to the one described in the preceding dictation task (3). Then they try to reproduce it without looking back at it, with the punctuation and capitals accurately inserted. They can share their results before checking with the original. Review Answer as many as you can of the following questions, and then check answers by referring back to the relevant passages in this unit. If you are working in a group, note down your own answers first alone, and then share with the other memebers of the group. Finally, check the answers together. What is writing? 1. Can you recall at least four aspects of writing that distinguish it from the other four skills (other than the fact that it is written!)? 2. What are the implications of these differences for teaching? A Course in English Language Teaching 165 11 Teaching writing Formal and informal writing 3. For what purposes might formal or informal writing be used? 4. Can you specify at least three important differences between formal and informal writing? Beginning writing: the letters 5. What are some problems in learning to write English for students whose L1 uses a different writing system? 6. Why is it important for students to be able to type as well as handwrite? Tasks that promote fluent writing 7. What kinds of tasks can be used to promote creative writing? And to give information? 8. Can you suggest two writing tasks that might be appropriate for elementary or intermediate classes, and two that are appropriate for more advanced ones? Process writing 9. What is ‘process writing’? 10. Suggest two problems associated with the giving of feedback on a preliminary draft, and then some solutions. Spelling and punctuation 11. How irregular is English spelling? 12. Can you suggest two activities that might help students practise punctuation? Further reading Ferris, D. and Hedgcock, J. (2005) Teaching ESL Composition: Purpose, Process, and Practice (2nd edn), Mahwah, NJ: Lawrence Erlbaum Associates. (A useful practical guide to the teaching of mainly academic students, providing guidance on stimulating, supporting and providing feedback on student writing) Hyland, K. (2003) Second Language Writing, New York: Cambridge University Press. (A particularly clearly written and accessible guide to the teaching of second-language writing) Kroll, B. (2003) Exploring the Dynamics of Second Language Writing, Cambridge: Cambridge University Press. (An interesting collection of articles covering both research-based theory and practical topics such as providing feedback) Shemesh, R. and Waller, S. (2000) Teaching English Spelling, Cambridge: Cambridge University Press. (Presents the basic rules of English spelling, with plenty of practical classroom activities to help learn them) Truss, L. (2003) Eats, Shoots and Leaves, London: Profile Books Ltd. (A ‘must read’ for the English teacher: the basic rules of punctuation entertainingly presented) 166 A Course in English Language Teaching 12 Assessment and testing Overview The sections in this unit are: 12.1 Functions and types of assessment. A general introduction to the topic of assessment, why and how we assess. 12.2 Giving a grade. Issues to do with giving a final grade at the end of a course. 12.3 Test design (1): testing accuracy. A list of test items, with critical discussion; paper versus computer-based tests. 12.4 Test design (2): testing comprehension and fluency. How listening, reading, speaking and writing can be tested. 12.5 Administering tests in class. Practical tips on the presentation and management of classroom tests. 12.1 Functions and types of assessment Functions of assessment The main reasons for trying to assess English proficiency are as follows: 1. In order to evaluate students’ overall level: for example, we may want to specify their level according to the Common European Framework of Reference for Languages (CEFR), or let them or their parents know how good their English is, or register them for an appropriate course. 2. In order to evaluate students’ progress: for example, we may need to give them a grade showing how much they have improved since the last assessment. 3. In order to evaluate how well students have learnt specific material during a course: for example, we may wish to assess how well they know a set of vocabulary items, a text or a grammatical feature. 4. In order to evaluate students’ strengths and weaknesses (‘diagnostic’ assessment): so that the students themselves can be aware of what they need to learn, and so that we can plan our teaching appropriately. We may get added benefits from assessment procedures. For example, we might learn some useful information about successes or failures in our own teaching, and they may help us to predict how students will progress in the future (‘prognostic’ assessment). Summative and formative assessment The most formal and prestigious types of assessment, such as state school- leaving exams, or international exams such as IELTS, Cambridge English: First or TOEFL, are summative in nature: they provide only a grade, often expressed as a A Course in English Language Teaching 167 12 Assessment and testing percentage, offer no specific feedback on aspects of performance, and are designed to summarize or conclude a period of learning. Summative assessment may be used as a basis for selection, as final school grades, or for acceptance into further education or employment. It may contribute little or nothing to ongoing teaching and learning; however, it is a part of the teacher’s job and we need to know how to do it effectively. Items 1 and 2 of the list above are used mainly for summative purposes. In contrast, most of the assessment that we carry out during a course (tests at the end of units in a coursebook, for example) is formative in nature: it may, like summative assessment, provide a grade in the form of a number, but it happens in the middle of a period of learning rather than at the end, provides clear feedback in the form of error correction and suggestions for improvement, and has the primary aim of enhancing future learning. For example, we might respond to a 6 dictation with a grade like 10 and the comment ‘You need to work on the spelling of the words I have underlined’ (corrective feedback). So the types of assessment described in items 3 and 4 above are essentially formative. Summative assessment is normally carried out either by the class teacher or by an external authority: a ministry of education, for example, or an internationally recognized body such as the British Council or Cambridge Assessment (ESOL). In contrast, formative assessment is almost always done by the class teacher as part of the teaching process, though it may be supplemented occasionally by the student’s own self-assessment (see below). Assessment tools Various tools are used to assess students’ language ability. Tests Testing is by far the most common basis for assessment. The criterion for success is a fixed level which the student is expected to reach (‘pass’); and the result is usually expressed as a percentage. Tests are relatively easy to design and check, take place at pre-set times and places, give clear-cut results, and are in general (rightly or wrongly) accepted as reliable bases for course grades. Tests also have useful functions in the course other than the assessment outcome. They act as ‘stations’ in the course programme, marking off the ends of units. They encourage students to review material in preparation. They are motivating, in the sense that students will work hard to do them well. They can give a sense of achievement and progress. Their content often provides some useful learning or review. In classes with discipline problems, they often provide a welcome oasis of quiet, concentrated work during the lesson. However, there are problems with tests as a basis for evaluation: 168 A Course in English Language Teaching  12 Assessment and testing They are not always valid (actually testing what they say they are). For example, a listening test based on long multiple-choice written questions may actually test reading rather than listening comprehension. They may not be reliable. For example, similar classes may get quite different results on the same tests because their teachers mark them differently. They are a one-off event which might not give a fair representation of the student’s overall ability. They discriminate against students with test anxiety who perform badly under test conditions. If they are the basis for crucial summative assessment in the student’s career, they can be extremely stressful for all students. Alternative assessment Other tools are available, sometimes termed ‘Alternative assessment’. These may solve some of the problems associated with testing, but they raise others. Teacher assessment. The teacher gives a subjective estimate of the student’s level. This is based on the performance of the student over time in a wide range of tasks, and it takes into account aspects such as the student’s effort and progress, or particular learning disabilities. On the other hand, it is inevitably subjective to some extent and may be seen by students, parents or school principals as unreliable. Continuous assessment. The final grade is some kind of combination of the grades the student received for various assignments during the course. Again there may be a problem of subjectivity, as the grades are given by the teacher. Also, the assignments and criteria for the different grades may vary from class to class, which makes it difficult to achieve standardization between classes. Self-assessment. The students evaluate their own performance, using clear criteria and grading systems. This is not very popular, even with students themselves, for summative assessment. And again there is the problem of subjectivity. However, for formative purposes, self-assessment can be very valuable, since it encourages students to reflect on and take responsibility for the evaluation of their own A Course in English Language Teaching 169 12 Assessment and testing learning. It is particularly helpful when it is combined with teacher assessment and discussed in a tutorial. Portfolio. The student collects examples of his or her own work over a long period to create a portfolio, which provides the basis for evaluation. This is a more student-oriented method of assessment, as although the teacher decides on the overall composition of the portfolio (for example, one essay, one test, one text comprehension assignment, etc.), the exact assignments that will go into it are chosen by the student. Portfolio assessment avoids the stress and ‘one-off’ problems of testing. It also provides a much broader basis for evaluation, though there is likely to be a disproportionate focus on writing. Portfolio assessment is widely used in secondary schools in Europe today. Its main disadvantage in practice is the amount of work for the teacher, who has to keep track of students’ work on the portfolios during the year to ensure they collect all the required components, as well as read and assess the portfolios of entire classes. Task What experience do you yourself have of any of these assessment tools, as teacher or student? What comments would you add to the definitions and criticisms shown above? 12.2 Giving a grade The most common practical problem relating to assessment that we face as teachers is how to decide the final grade of a student, whether it is at the end of a course, at the end of a term or at the end of a year. Criteria The first decision to be made is what standard you will use to judge your students’ level: whether you are going to use criterion-, norm- or individual-referenced assessment. Criterion-referenced assessment means that you judge the student according to some fixed criterion. This can be based on an estimation of what it is reasonable or desirable to demand from students according to their age, career, level, stage of a course, etc. The criteria might also be based on the levels of the CEFR, for example. Norm-referenced assessment means that you evaluate the student’s performance relative to what you would expect from the particular group. In this case, a group of less advanced, or learning-disabled, students would be assessed according to different standards from those applied to an advanced student within the same school. Individual-referenced assessment means that you relate the assessment of an individual student to his or her own previous performance, or to an estimate of his or her individual ability. You would give a student a high grade if he or she has worked hard and made impressive progress, even if by norm- or criterion- referenced standards the student might have received a fairly low one. 170 A Course in English Language Teaching  12 Assessment and testing Components of the grade Having decided what your criterion (or combination of criteria) will be, you then need to decide what information you will use as a basis for the grade. You may not have much choice: some schools have rules for their teachers about what the final grade must be based on. But if you have a choice, then it is probably best to take into account ongoing work as well as tests: whether a student has made an effort and progressed, whether he or she has consistently submitted homework assignments, has assembled all the components of a portfolio satisfactorily and so on. In some classes, particularly the younger ones, you may also want to include behaviour as a component: whether or not the student has been punctual, attentive and cooperative. The use of such components in a grade is fairer than assessing students only on the basis of a one-off test, which may not, for reasons given earlier, provide a fair sample of what they can do. It also helps student motivation. This may not be very ‘educational’, but it is a fact of life: if we know that something is going to affect how we are assessed, then we are more likely to make an effort than if we know it is not. For example, if students know that completing homework assignments throughout the term accounts for 10% of their grade, they are more likely to do them. We need to decide what weighting (percentage of the final grade) we will give to the different components: which means, of course, rather more work for us than just copying out the results of a test. For example: Final test or portfolio – 50% Class tests – 10% Ongoing class work and attentiveness – 10% Homework assignments – 10% Personal effort – 10% Overall progress since last assessment – 10% In different situations, it might be appropriate to remove or change some of these criteria, or add others, or change the weighting. Your own pedagogical approach and student expectations will also make a difference. Task Think of a class you have taught or observed (or, perhaps, one you have been a student in). Do you think the components and weighting shown above would be suitable? If not, how would you change them? How the grade can be expressed How do you express the grade you give the student at the end of the course? There are various possibilities: Percentages are probably the most common, though in different places the actual value placed on the different percentages may vary. For example, in some places 40% is a pass, in others it is 60%; some student populations consider 75% a high A Course in English Language Teaching 171 12 Assessment and testing grade, others would think anything below 85% unsatisfactory. So you would need to find out how the different percentages are regarded at your school. Letters, words or phrases, such as ‘A’ or ‘B’; ‘Good’, ‘Very good’ ‘Excellent’ look a little less impersonal, less definitive than percentages; however, the students and other stakeholders (e.g. parents, employers) often ‘read’ them as definitive number-type grades, exactly as they read percentages. Profiles are a totally different kind of expression of assessment, comprising a number of separate grades on different skills or sections of knowledge, so that there is a possibility of describing the performance of an individual student in more detail, showing his or her various strengths and weaknesses. So you might, for example, provide a rubric with categories such as reading, writing, speaking, listening, grammar, vocabulary, and give a grade, or remark, for each. This provides a more rounded view of the student’s level, but obviously involves a lot more work for the teacher. Evaluative comments, such as ‘Well done! You have worked hard’, without any expression of level of achievement, avoids the difficult and sometimes unpleasant job of actually having to give a grade; however, the institution will normally demand a grade, as will other stakeholders, such as parents. Students also, in my experience, want to see a grade: they need some kind of clear-cut evaluation of how well they are doing. So it is probably best in most contexts to give an indication of achievement through a grade, but accompany it with encouraging and constructive comments. Task Consider the way grades are awarded and expressed in a teaching context you are familiar with, either as a student or as a teacher. What would be your criticisms (positive or negative)? What improvements can you suggest? Practical tips 1. Tell students early on what your criteria are. Right at the beginning of the term or course, make sure you explain to the students on what basis the grades are given: whether you are taking into account aspects of ongoing work, or only results of a final test. Don’t leave it to the last minute: students should know from the start how they are going to be assessed. 2. Discuss the grade with individual students. If your class is not too big, and if time allows, try to set up individual meetings with students. Give them general feedback on their performance, tell them what grade you intend to give them, ask them what grade they consider they deserve; clarify and discuss any differences. You may sometimes change the grade after this consultation. This will remove the stress of not knowing what the final grade is until they get it in writing; and in some cases it may help you decide on a fair grade. If you are short of time, the meetings can be done in class time while the rest of the class is doing individual work. 3. Keep grades private. Don’t make the grades public (unless your institution insists on it). Students usually prefer to be told privately what their grade is, and then choose themselves whether, and with whom, to share it. 172 A Course in English Language Teaching  12 Assessment and testing 12.3 Test design (1): testing accuracy There are various types of test items for checking accurate usage of vocabulary, grammar, spelling and punctuation. You may already have come across some of them if you have read Unit 5: Teaching vocabulary or Unit 6: Teaching grammar. In this section you will find a list of the most common types of test items, with some notes on their advantages and disadvantages. For testing techniques for the assessment of listening or reading comprehension, or of spoken or written ability in general, see Section 12.4 below. The test items below are normally used in conventional class tests given at the end of a teaching unit, term or year. They usually appear within a written test which includes several sections using different types of items. Occasionally some of these test items may be used in an oral test, such as an interview. This provides the tester with information about how well the student can understand and produce the spoken forms. Interviews are, however, rather time-consuming as well as expensive, since testers need to be paid for their time. So schools with tight budgets tend to use written tests. Some test items can be administered, answered and graded using the computer: but these are normally only ones that have a limited set of clear, predetermined right answers, such as multiple-choice, matching or gapfilling. See Computer testing on p. 178 below. We need to take into account various considerations when selecting items to use for a particular test. What will it tell me about the student’s knowledge? For example, will it reveal the student’s ability to produce the item, or just show that they understand it? Will it provide evidence that the student can use the item in appropriate contexts or not? How easy is it to compose? Will it take me a long time to think up and write out the item? Is there a source (the coursebook, a website) which will provide me with ready-made items I can use? How easy is it to check? Does it require only a quick, objective check based on a single possible right answer, or will I need to use my own judgement in assessing the answer? Can it be used in the design of a computer test? Are the answers clear and objective enough to enable it to be administered and checked through a computer? The test items that are listed below are divided into two groups: those which are conventionally used quite a lot and which you are likely to be familiar with; and those which are used less but have various useful functions that are worth considering. A Course in English Language Teaching 173 12 Assessment and testing Task Choose five test items from the list below, and comment on them based on the questions above. Then read on. Frequently used test items 1. True/false. This may either be in the form of a statement (‘true’ or ‘false’), or as a question (‘yes’ or ‘no’). Underline ‘True’ or ‘False’. a) London is the capital of France. True / False Write ‘Yes’ or ‘No’. b) Is London the capital of France? _________ 2. Multiple-choice. There is one correct option out of (usually) four. Circle the right answer. A person who writes books is called a) an engineer. b) an accountant. c) an author. d) a baker. 3. Gapfill. The ‘base form’ of the word that is to fill the gap may or may not be provided; the placing of the gap may or may not be shown. Complete the sentences. a) They ___________ (go) to Australia in 2007. Or b) The money was _____________from the bank. (steal) Or c) ________ you like action movies? Or d) I’ve seen that film. (never) 4. Matching. Each item is to be matched with one other. Match words that mean the same. a) large small b) unhappy many c) a lot big d) little sad 5. Dictation. The tester dictates a passage or set of words; the student writes them down. 6. Focused cloze. The target lexical or grammatical items are omitted from a passage. 174 A Course in English Language Teaching  12 Assessment and testing Insert words in the past tense. Beowulf 1. was (be) a great warrior 1000 years ago. He 2. ____________(win) many battles against monsters and dragons. Grendel ____________ (be) a terrible monster from Denmark. He 3. ____________ (have) big teeth and he was very strong. Grendel 4. ______________(not sleep) and 5. ______________(not eat). He 6. ____________(drink) blood. …1 7. Transformation. This usually involves alterations such as changing the tense or voice (active/passive), or number (singular/plural). Put into the past tense: I go to school by bus. 8. Rewrite. A sentence is to be written that paraphrases the one provided. Complete the second sentence so that it means the same as the first. He came to the meeting in spite of his illness. Although … _______________________________________________ Less frequently used test items 9. Sentence completion. The student may complete the sentence any way he or she likes provided the language is acceptable. Complete the sentence. She will come to the party if … 10. Translation. This may be either from the L1 into English, or from English into the L1. 11. Mistake correction Correct the mistake. *We talked to the man which is in charge of the project. 12. Wordsearch. The target items are hidden within a criss-cross of words within a grid, to be identified by the student. 13. Unscramble sentences. The sentences have words in jumbled order: the student has to write out the words in the correct order. 1. Davis, F. and Rimmer, W. (2010) Active Grammar 1 (p. 27). Cambridge: Cambridge University Press. A Course in English Language Teaching 175 12 Assessment and testing Comment A characteristic of all the more frequently used test items is that they are easy to check. Most are also fairly easy to compose, with one notable exception (multiple- choice, see below). 1. True/false items check only receptive knowledge: the fact that students got the answer right does not indicate that they would be able to produce the target item themselves correctly or appropriately. And note that they have a fifty-fifty chance of getting it right even if they are guessing. 2. Multiple-choice may be used for the same testing purposes as true/false items and checks knowledge more reliably since it offers more options (there is only a 25% chance of getting it right by chance). Good multiple-choice questions, however, are surprisingly difficult to design. They often come out with more than one possible right answer, or no clear right answer at all, or one over- obvious right answer. Also, the punctuation and aligning of the ‘stem’ and ‘options’ can be tricky. Finally, less experienced test-writers tend to make the right answers the longest ones, which may give them away to students who are aware of this tendency. 3. Gapfills, again, test mainly receptive knowledge. You need to be careful to design a gapfill item so that there is only one right answer, or a very limited number of right answers, otherwise it becomes difficult to check. For this reason it is generally advisable to limit choice by including a root word in parenthesis or in a ‘word bank’. 4. Matching can be used not only to elicit knowledge of synonyms, as shown in the example, but also for opposites, collocations and grammatical cohesion (appropriate sequence of tenses, for example). Design is not too difficult, but as with multiple-choice questions, you do need to take care that there is only one right ‘match’ for each item. 5. Dictation mainly tests spelling, sometimes punctuation and, perhaps surprisingly, listening comprehension (people can usually write things down accurately from dictation only if they understand them). It may supply some information on students’ receptive knowledge of pronunciation, grammar and vocabulary. Note that when checking, you may have a problem deciding how much weight to attribute to different mistakes. If you are teaching a monolingual class whose language you know, you might use ‘translation dictation’: dictate a word or short text in the L1, the students write down the translation. But here again, there may be different possible right answers which you will need to take into account when checking. 6. Focused cloze can be used to test grammar, vocabulary, spelling or punctuation, depending on which items you choose to delete. Marking can be slightly less straightforward than for previous items: you may find it difficult sometimes to decide if a specific item is acceptable or not. If you insert multiple-choice options at each gap, composition is more time-consuming, but the marking becomes easier. 7. Transformation items are still in use, though less popular than they used to be. Their validity may be suspect: they check the ability of the student to 176 A Course in English Language Teaching  12 Assessment and testing produce correct forms as transformations (to write, for example, went as the past of go), which is not the same as testing productive knowledge of grammar. A student may perform well on transformation items without knowing the meaning of the target structure or how to use it in context. 8. Rewrite tests the same sort of thing as transformation. However, it is likely to reflect more thorough knowledge of the target items since it involves paraphrasing the entire meaning of a sentence rather than just transforming a particular item. As with the focused cloze, you may occasionally find it difficult to decide if a specific answer is acceptable or not. The less frequently used test items are not so popular for a variety of reasons. 9. Sentence completion is difficult to check, since there is often a very large number of possible right answers. Its big advantage is that it tests production: shows whether the student can produce correct sentences and use the target items in appropriate (though limited) contexts. 10. Translation is still frowned on by some teachers and methodologists, though far less than it was a generation ago. It is actually a useful technique in a monolingual class whose teacher also speaks the students’ L1. The translation of a language item to or from English can give very quick and reliable information about what the student does or does not know, particularly when it involves entire units of meaning (phrases, sentences) within a known context. Items are fairly easy to design and may be more, or less, easy to check depending on how close and obvious the translations are. 11. Mistake correction is, again, something which many teachers feel uncomfortable with. There is always the worry that exposing students to a mistake in print might reinforce that mistake rather than correcting it. If used, it has to be very clear what the mistake is. One possibility is to cross out the wrong words (for example, ‘*We talked to the man which is in charge of the project’) so the students only have to decide what words should replace them. Another is to write an ‘insert’ symbol ^ where something is missing. But both design and checking are fairly straightforward. 12. Wordsearch. This is a fairly motivating test item for teenage or younger learners, though not so good for adults. It is a useful item to put at the end of a test to keep the faster workers busy while the others are finishing. But of course it does not provide information on the students’ ability to produce the item or use it in context, and may not even tell us if they know the meaning; and it is extremely difficult for students with some forms of dyslexia. It can be produced and checked through computer programs (e.g. Word Search Maker: http://puzzles.about.com/od/wordsearches/tp/word-search-makers.htm). It is relatively easy to check. 13. Unscramble sentences. Most of the same positive and negative criticisms that apply to Wordsearch apply here too; though it is less attractive to learners and rather more difficult. It is particularly useful when testing aspects of grammar where word order is crucial – in questions, for example, or other structures that demand inversion. A Course in English Language Teaching 177 12 Assessment and testing Computer testing The computer is very good at testing receptive knowledge and mechanical knowledge of forms (e.g. irregular past tenses) through dual- or multiple- choice, gapfills and matching items. There is online software for creating your own tests with those exercise types: see, for example, the Hot Potatoes website (http://hotpot.uvic.ca/). There is also a large number of ‘test yourself’ sites online, where students can test their own knowledge and get instant feedback: for example, Easy English (www.easyenglish.com/), or Learning English – Quizzes (www.bbc.co.uk/worldservice/learningenglish/quizzes/). Computer software, however, is not yet able to assess students’ ability to use the target language items to express themselves in longer, more open responses, such as sentence completion. For this kind of assessment, there is at the moment no alternative to tests composed (or adapted) and assessed by the teacher. Practical tips 1. Clarity. Make sure the instructions for each item are clear. They should usually include an example with its solution. For low-level, monolingual classes, it may be appropriate to have the instructions in the students’ L1, as well as in English. 2. ‘Doability’. The test should be quite doable: not too difficult, with no trick questions. It’s sometimes difficult to judge the doability of your own tests, so it’s worth asking a colleague to read through and check they can answer the questions before you give the test to students. 3. Marking. Decide exactly how you will assess each section of the test, and how many points you will give each out of the total. Make the marking system as simple as you can, and inform the students what it is: write in the number of points assigned to each section on the test sheet itself. 4. Interest. Try to choose interesting content and tasks in order to make the test more motivating for the students. 5. Varied level. Lower-level students should feel that they are able to do a substantial part of the test, while the higher-level ones should have a chance to show what they know. So make the earlier items fairly easy, and define one or more of the more difficult ones as optional. (For more discussion of tests and worksheets for mixed-level classes, see Unit 19: Learner differences (2): teaching heterogenous (mixed) classes.) 6. Occasionally, let students compose their own tests. Students can be told exactly what you intend to test, and then write their own test items, individually or in groups. You then collect these items, correct them if necessary, and use some or all of them as a basis for the test. This is in itself an excellent review of the test material and also reduces test anxiety. 12.4 Test design (2): testing comprehension and fluency Tests that assess comprehension through listening and reading, and fluency and accuracy in speaking and writing, rather than knowledge of specific language items, need a different set of testing techniques. 178 A Course in English Language Teaching  12 Assessment and testing Listening comprehension This is usually tested as a skill on its own, although in real life it normally occurs in conjunction with speaking (for discussion of the teaching, rather than testing, of listening skills, see Unit 8: Teaching listening). A test which involves both speaking and listening is the interview, described under Speaking later in this section. Dictation and repetition. As mentioned previously, a student can only normally write down more than a word or two accurately from dictation if he or she has understood it. The same applies to oral repetition. So one way of assessing listening comprehension at the most basic level is to ask students to repeat what they have heard, either in writing or orally. They should, of course, hear the source text more than once. Text + comprehension questions is probably the most common form of listening comprehension test. The student hears a text, usually two or three times, and is asked to answer questions on it. The questions may relate to gist or details of content. For convenience of checking, they are very often multiple-choice. See, for example, the listening task shown on p. 54, which is a fairly typical test format. This kind of test can easily be computerized using audio texts followed by questions with a limited possible set of right answers. Taking notes is a useful test of listening comprehension, but one that demands, of course, the ability to write quickly and clearly as well as understand what is heard. It is a component of some international examinations, such as the Cambridge English: Key and Cambridge English: Preliminary. Reading comprehension Ideas for teaching students to cope with reading texts have been discussed in Unit 3: The text, and the promotion of reading fluency in general in Unit 10: Teaching reading. This section deals only with tools for assessment of reading comprehension. Reading aloud of single words can, obviously, show only that the student can decode the letters accurately; however, reading aloud of a text with appropriate prosody (pause, intonation, stress) can be done well only if the text is understood. So reading a text aloud is a very easily designed and administered test of reading comprehension, in very much the same way as dictation and repetition are tests of listening comprehension. It is, on the other hand, time-consuming because it has to be done through one-to-one interaction with the teacher. The student should usually have time to read, reread and prepare before such a test: only at very advanced levels can we expect students to sight-read competently. Text + comprehension questions, as with listening, is the most common format of the reading comprehension test. The students study a text and answer questions, which are commonly gapfills and multiple-choice but may also be open, inviting the students to respond in their own words. Cloze is another way of checking general reading comprehension. Unlike ‘focused cloze’, where specific target items are deleted, the reading compre- hension cloze test normally has words deleted at regular intervals (commonly every seventh word), so it is unpredictable whether these will be grammatical or lexical items. However, it is possible to be flexible with this point. A Course in English Language Teaching 179 12 Assessment and testing Jumbled paragraphs. Students are given a text with the paragraphs in the wrong order, and they have to sort them out. Their success depends not only on their comprehension of the content, but also on their awareness of the typical discourse structure of the genre and of cohesive devices (use of pronouns, connectors such as however, moreover, discourse markers such as first, on the one hand, finally). Speaking The assessment of a student’s ability to speak fluently, accurately and appropriately is particularly problematic for two reasons. One is practical: unlike the other three skills, speaking can only normally be tested in individual (or, occasionally, pair- or small-group) interaction. This means that it takes a long time to test an entire class, and it is expensive to pay the testers. The other problem is reliable assessment: there cannot possibly be ‘one right answer’, so there is no possibility of objective or computer-based grading. Moreover speech is fleeting; it cannot usually be ‘reread’ and reconsidered for assessment. It is also difficult to retain speech in the memory long enough to assess its level. It can help to record the student’s speech and then listen to it again later, but this increases the time and expense even more. The use of scales of standards, or rubrics can help to solve the problem of reliability by making sure that there are clear criteria for the different possible grades. Here is an example, assuming that the test is based on an interview (see below). The following are some common formats used in oral testing: 1 Does not speak at all 2 Single-word responses to interviewer cues; difficult to understand. 3 Communicates in brief, hesitant phrases; often difficult to understand; only speaks in response to cues from interviewer. 4 Uses short sentences, and communicates limited messages; can give slightly longer responses to interviewer cues. 5 Can use longer sentences and convey clear messages; occasional hesitations; able occasionally to show initiative and prolong an exchange with interviewer. 6 Speaks fluently and clearly; engages in a cooperative dialogue with interviewer; takes initiative. Interview consists of a conversation between two people. It is the most common context for speech in daily life and therefore should provide useful and reliable evidence of the ability of the student to converse in English. It should be noted that it does in fact test listening comprehension as well as speaking. Its main disadvantage is the problem of attention for the interviewer who has to initiate and maintain a conversation as well as assessing. With an unwilling or shy interviewee, this can be very difficult. The problem can be avoided by having one person interview and another assess, but the expense of two testers is not acceptable in most places. 180 A Course in English Language Teaching  12 Assessment and testing Picture description is particularly suitable for younger learners or beginners. The student describes a picture or a series of pictures. It is easier for the tester, who does not have to initiate conversation and can devote his or her attention to assessing the student’s performance. Presentation is a longer, more advanced procedure. The student is asked to present an extended description, explanation or other oral account, while the tester simply listens and assesses. The problem here is that students may memorize their presentations in advance, which gives no idea of their ability to compose and deliver spontaneous speech. An alternative is to give the student a topic, two minutes to prepare, and then ask them to speak about it for a minute or two. Group or pair discussions can be used in order to assess the speaking ability of two or more students at the same time, while the tester simply sits at the side and listens. This possibly saves time, but there is always the chance that the less assertive students, however good their spoken English, may not get sufficient opportunities to speak. Discussion needs to be skilfully led to ensure that everyone gets a reasonable chance to participate. Writing A written assignment used for assessment purposes can of course be done on a computer, in which case students can use computer tools to check their spelling and grammar. Some teachers prefer not to let students use computers when doing writing tests for this reason. However, in my opinion, the use of such tools cannot disguise poor writing ability; and in any case since much, if not most, English writing is now done with these tools available, it does not make sense to exclude them from tests. A trickier problem with using computers is the possibility that students will copy-paste passages – or, indeed, full essays – from the Internet. However, it should be possible to make sure that your students do not have internet access. You can usually tell if the writing is not at the same level as writing assignments you have received from the same students previously; and they should also be aware that whole texts copied from the Internet can usually be fairly easily identified through search engines. Assessing free writing is very difficult. We do not have the problem of the fleeting nature of the input, as in speaking, but we do need to assess various aspects: accuracy, coherent organization, content and so on. How much weight should be given to each? The suggested rubrics shown on the next page are adapted from a variety of assessment scales from different sources (e.g. Weir & Roberts, 1994) and are appropriate for writing at an intermediate or advanced level. Compositions are probably the most common form of tests of writing skills at intermediate or advanced level. The student is given a topic, or sometimes a genre, communicative purpose and target audience (see, for example, the set of writing tasks listed in Unit 11: Teaching writing, p. 158), and asked to write a composition of a set length. Brief descriptions and dialogues can be used to test writing at elementary level. The student is given a picture to describe, or the beginning of a dialogue to