learning strengths
97
C H A P T E R F O U R
Assessing language use through tasks
The primary aspiration in language teaching programmes, for most parents and teachers, is that their children will become effective language users, that is, they will be able to communicate in the target language. When children are language users they are able to use the target lan- guage to exchange meaning in ways appropriate for the purpose and context. This may happen when they are interacting orally with their teachers, their peers in the classroom or others outside school, or when they are listening to and understanding their teacher talk, and when they are listening to stories or viewing videos. Language use takes place when children are reading stories and information texts and understanding the events, ideas and information in what they read. When children are producing their own writing in the target language, however simple the writing may be, they are using language.
In some teaching situations where external testing exists, parents and teachers also hope that their children will pass the required tests. Most parents and teachers would expect that tests promote their chil- dren’s ability to use the language, though this is not always the case. Since external language tests can influence strongly the nature of language teaching and learning in the classroom, the alignment of formal tests with language use is highly desirable, as I discuss further in Chapter 9. Language educators can ensure that children learn to become language users by giving them the kinds of learning opportunities and conditions I described in Chapter 2. Alongside this, assessment can be structured so that it supports the development of language use; this is done by assessing primarily through language use tasks. Language use tasks give
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
teachers and assessors information about the child’s ability to use lan- guage in communicative ways.
Recent developments in performance assessment have provided new directions in assessment that inform the approach to assessment through language use tasks. Therefore the first section of this chapter looks at the assumptions and characteristics of performance assessment. Further principles of effective task-based assessment of young learners follow. These assumptions underpin the approach to assessment adopted in this book. Through language use tasks children have the opportunity to show their ability to use language, exchanging meaning according to their own purposes, and in spontaneous ways according to the context.
Principles and frameworks for the selection of language use assess- ment tasks are needed, whether for the classroom or for external tests. How do teachers and assessors select the best assessment tasks for young learners? What kinds of assessment tasks give children the best learning opportunities, as well as the best opportunity to show what they can do? Wrongly selected assessment tasks may disadvantage some children. Some children may require support during tasks – are there ways that assessment tasks can be analysed beforehand to ensure that adjustments can be made to ensure the child’s best performance? Principles and frameworks for the selection of assessment tasks are provided in this chapter before we move on to ways of assessing in the classroom in the next chapter.
Performance assessment
I use ‘performance assessment’ here as an overall term to refer to a family of like-minded assessment approaches including ‘alternative’ and ‘authentic’ assessment (see, for example, Herman, Aschbacher and Winters, 1992; O’Malley and Valdez Pierce, 1996). Performance assess- ment refers to assessment that ‘involves either the observation of behavior in the real world or a simulation of a real-life activity’ ( Weigle, 2002). In these approaches, assessment through selected-response items is avoided. The item in Figure 4.1 in which children are asked to select the right word is a selected-response item. It is also called a dis- crete-point assessment item because it is intended to measure one point of language knowledge only (in this case knowledge of accurate use of personal possessive pronouns).
98
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 99
In performance assessment teachers tend to avoid using assessment items like the one above that specifically target language for its own sake. Rather, performance assessments give children opportunities to use the language for real purposes, and in real or realistic situations, and assess their attempts to do so successfully. Children’s grammar and vocabulary knowledge is assessed as part of their performance in real-world or real- istic tasks rather than separately in discrete-point assessment items. Teachers can observe and evaluate the performance as a whole (did they achieve the purpose of the task?) and also the elements of language use, such as vocabulary and grammar, within the performance in the task (to what extent did they use a range of vocabulary? To what extent was the performance accurate?). Performance assessment involves teachers and assessors making decisions on performance by checking performance against criteria, rather than by comparing students’ performance against the average performance of all learners. Ways to do this in language assessment are described in Chapter 8.
The principles of performance assessment also stretch beyond assess- ment into the teaching and learning process. The assumptions and char- acteristics of performance assessment have been summarized by one writer as follows:
• Students are active participants rather than passive subjects • Evaluation and guidance occur simultaneously and continuously • Processes as well as products are evaluated • Development and learning need to be recognized and celebrated • Multiple indicators and sources of evidence are collected over time • Results of the assessment are used to plan instruction, improve
classroom practice, and optimize children’s learning • The assessment process is collaborative among parents, teachers,
children, and other professionals as needed ( Jalongo, 2000, p. 287)
Fill in the blanks with the correct words from the box. You may use the words more than once
She has a dog. She likes to play with _______ dog. I have a storybook. Would you like to borrow _______ book? He has a canary. He takes good care of _______ bird.
my her his our their she your
Figure 4.1 Example of discrete-point assessment: a fill-in-the-blank task.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Thus, performance assessment brings with it both a focus on children’s abilities in real-world tasks, but also attention to broader characteristics of assessment that encourage amongst other things active participation, attention to the processes of learning, and involvement of parents and children in the assessment process. The various assumptions and char- acteristics of performance assessment underpin the approach to assess- ment outlined in this book. Young learners learn best through activities that are concrete and meaningful, and evidence of their language learn- ing is most likely to be present in language use assessment tasks that have similar characteristics to those in the child’s real world. Performance assessment has had a major influence on thinking in assessment in recent years; this influence has been particularly strong in classroom- based assessment where there have been more opportunities to imple- ment some of the ideas, than, for example, in external testing. However, there continues to be a push to incorporate characteristics of perfor- mance-based assessment into more formal assessment situations, including large-scale testing.
Language use tasks
We will now look more closely at a definition and examples of language use tasks. Tasks, traditionally discussed as teaching activities with an intended pedagogical purpose (Purpura, 2004) have more recently been character- ized by their ability to elicit interaction and negotiation of meaning and to engage learners in complex meaning-focused activities (Nunan, 1989, 1993; Berwick, 1993; Skehan, 1998). This emphasis on the communicative goals of tasks is taken up in the definition of a language use task used in this book. A language use task is ‘an activity that involves individuals in using language for the purpose of achieving a particular goal or objective in a particular situation’ (Bachman and Palmer, 1996, p. 44). Language use tasks are goal-oriented, meaning that the learner knows what is to be achieved in the task, and they are specific to a particular situation. Each instance of language use is virtually unique (Bachman and Palmer, 1996, p. 44). Language use tasks can involve listening, speaking, reading or writing and may entail a combination of these activities.
I interpret and expand this definition for young learners. In language use tasks, children’s language participation involves a degree of spontaneity and creativity; they make their own meaning, producing meaning or com- prehending meaning, according to the purpose and the requirements of
100
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 101
the situation. The creativity and spontaneity rests in children drawing from their ‘language resource’, that is, from the language and language rules they have internalized. These may be chunks of unanalysed language or new rule-based constructions (see Chapter 2). Children use this language to fulfil a communicative purpose and do so in an appropriate way for the language use context. Language use tasks can take place in a very simple and supported way, or in a more extended, complex and independent way. We do not necessarily expect creative language use in language use tasks to be accurate, wide in vocabulary use or appropriate, but we expect to see growth in these features as experience grows. Language use tasks do not necessarily need to be noisy or time-consuming (characteristics that are avoided in some teaching situations).
The example of a language assessment task for beginning young lan- guage learners in Figure 4.2 involves children filling in blanks in sentences helping them to write about a story they have heard. They are asked to use their own words to fill in the gaps. The task is different from the discrete- point item in Figure 4.1 above because it has gaps for children’s own lan- guage. Children can write the story they have heard in their own words, with support from the part-sentences supplied, and the teacher can expect to find some spontaneous language use in the gaps. This simple task requires children to write a story (the purpose) in language appropri- ate to a narrative (the situation). The task could be left more open, with children being able to choose what happened in the story.
The example of a language assessment task shown in Figure 4.3 is for more advanced language learners. Children are asked to write what hap- pened in a science experiment they have observed. The questions give them guides for the task and, at the same time, provide a learning struc- ture for a procedural genre.
Children are given a stapled booklet with eight pages, with a simple story written by the teacher through the pages. There are blanks in the sentences on some pages. Children are asked to fill in the blanks, and to draw a picture for each page, illustrating what is happening in the story. The children have already heard the story: it has been read to them several times, and talked about in classroom activities.
Page 1: The little girl’s name was Jane.
Page 2: One day she found a .........................
Page 3: She felt very ......................... about this.
Page 4: Then she .........................
Figure 4.2 Example of a language use task for beginning young language learners.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
102
Language use tasks in the classroom
There is virtually an infinite number of language use tasks that can be used in assessing young language learners. Many classroom language teaching tasks can be used for assessment. Teachers might observe chil- dren’s performance through the task, carry out on-the-run assessment as they teach, or set up the task for formal assessment with criteria made known to the children at the beginning of the task. On-the-run assess- ment is that which is integrated into the busy-ness of teaching (see Chapter 5). The following list of classroom language use tasks ( Table 4.1) from Williams (1994) evokes the young learner language classroom and illustrates the characteristics of tasks that are suitable for young language learner assessment. Many, if not all, of the teaching and learning tasks used in the classroom, or parts of them, may be suitable for assessment. While these tasks are aimed at children at the lower end of elementary school, adaptations of these, moving into more content-based instruc- tion, would be suitable for learners in upper elementary school.
Teachers are able to select from a range of different language use tasks, depending on the proficiency level of the children and their interests and the demands of the curriculum. Tasks can involve problem-solving, they can be information gap tasks (where children need to find out informa- tion to complete the task), opinion gap tasks (where children have to find out someone else’s opinion in order to complete the task), affective gap tasks (where children have to find out what others are feeling to complete the task), games, drama tasks, tasks using pictures, literature-based tasks,
Children have observed a simple science experiment. They take notes. They are then asked to write what happened in the science experiment. They have been given a paper with headings to guide them.
Name: ____________________________________________________
What were we trying to find out?
Describe the equipment that was used.
What did we do?
What happened?
What did we find out?
Figure 4.3 Example of a language use task for more advanced young language learners.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 103
and writing tasks to name a few (Scarino, Vale, McKay and Clark, 1988, Book 4, p. 29). Many games and drama tasks are suited for classroom assessment where the teacher observes and notes children’s performance as the rhythm of the task proceeds. Personalized assessment tasks are suitable for young learners – that is, tasks where the topic is related to their own interests and lives. Expressive writing about oneself and one’s family and friends, questionnaires and surveys, individual interviews about feelings and ideas are examples of personalized tasks. Literature- based tasks are very suitable for all learners (Falvey and Kennedy, 1997). Literature-based tasks generally use stories that children have been reading. Children may, for example, read stories aloud (assessing their
Table 4.1 Examples of classroom language use tasks for young learners ( Williams, 1994, p. 209)
Doing puzzles and Writing and Using maps solving problems solving riddles
Measuring and weighing Conducting surveys Growing plants things (e.g., food, birthdays,
traffic surveys)
Following and Interviewing people Making things (e.g., writing recipes (e.g., parents, people witches, spacemen,
in the neighbourhood, stranded on an island) different occupations)
Inventing and designing Planning things (e.g., Inventing games (e.g., things (my ideal . . . an outing, a party) board games, writing A machine to . . . fashions) the instructions)
Choosing (e.g., films, Writing letters (for Reading and designing clothes) real purposes) brochures
Designing and recording Finding out (e.g., what Filling in forms a TV programme things are made of, what
materials are used for, how things grow, whether objects float or sink)
Studying the local Making charts and Using songs and rhymes environment (e.g., plants, graphs birds, buildings)
Listening to stories Painting, drawing and (a particularly motivating talking about what form of language input, we are doing and recommended as a daily activity)
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
104
ability to read this level of text aloud), draw pictures based on a part of the story (assessing various constructs including comprehension of the sequence of events in the story, comprehension of description), write questions about a story or a poem (assessing comprehension as well as ability to write questions), complete an unfinished story or play (assess- ing comprehension of the story so far, as well as ability to write and to write creatively), or answer questions about the story (assessing different levels of comprehension depending on whether the questions are literal or interpretive). The following task is a simple writing task asking children to read a poem and answer a question.
What is the wizard doing in bed? Write down what you think.
Figure 4.4 An example of a literature-based response (Poem from Foster and Lewis, 1996)
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 105
Classroom teaching affords many opportunities for language use assessment to be embedded in teaching tasks. In Cameron’s (2001) teaching task framework in Figure 4.5 there is at least one language use assessment task (see teaching activity in bold) that gives the teacher opportunities to assess children’s language use. In this assessment task, early language learners are asked to write their own sentences about Hani’s weekend. Children will be putting into action the language they have recalled and practised in the teaching task. There are also on-the- run assessment opportunities to observe children’s developing vocabu- lary and grammar, and their ability to construct the sentences that are being practised.
Language use tasks therefore give teachers opportunities in the class- room to assess children’s ability to use language. Chapter 5 deals in more detail with language assessment in the classroom.
TASK Say sentences about Hani’s weekend
Preparation CORE Follow up
Language Activate previously Oral production Written production learning goals learnt lexis. of sentences of Hani’s sentences.
Practise past from grid. Composition of own forms of verbs. sentences.
Teaching Teacher-led: (1) Whole class (1) Teacher writes activities (1) Use of single introduction key words on
pictures to of grid and board, next to prompt recall teacher pictures. of lexis. modelling (2) Teacher models
(2) Divide board of sentences writing sentence into two and (2) Pair from grid. recall/practise production (3) Pupils write past forms. of own own sentences
(3) Pairs practise sentences about Hani’s with single about Hani’s weekend. pictures. weekend e.g. (4) Pair checking
P1 points to of accuracy. a box and P2 says sentence.
Figure 4.5 An example of an embedded language use assessment task in a classroom teaching task (Cameron, 2001, p. 34)
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
The place of selected-response and limited-response tasks in young learner language assessment
In selected-response tasks, children are expected to select a response from input which may be language (spoken or written) or non-language (e.g., pictures). Selected-response tasks include multiple-choice items, picture cloze and picture-matching vocabulary items, where children are given input (lists of words, pictures) from which to select. Limited- production tasks are those in which only a limited response is required, usually a word or a sentence. Limited-production items include gap-fill (where children fill in a gap in a sentence), fill-in-the-blank items and short-answer tasks. Selected-response and limited-production tasks are often used with young learners because of their limited language proficiency, and because they need support within the task to under- stand what is required and to be able to complete the requirements of the task.
Selected-response items and limited-production items may be language-use oriented or they may be language-item oriented. This dis- tinction is important for those involved in the assessment of young learn- ers. In the example of a language-use oriented selected-response task used in the Cambridge Young Learner Test in Figure 4.6, children are required to listen to a tape-recording of a description of a scene at the beach. They have a picture in front of them showing children and adults playing, doing different things (playing with a ball, rowing, eating an ice cream and sitting under an umbrella). A list of names accompanies the picture. The children are asked to draw a line from the name to the action. This item involves the specific ability to listen for detail. The item has been checked with children for its age-appropriateness. Children are using language when they listen to the input, look at the picture and make a decision about who is doing what.
Language-item oriented selected-response and limited-production tasks need to be used with more care with young learners. Examples of these are gap-fill items (She is _____ some water to make a cup of tea. (1) baking (2) steaming (3) boiling (4) roasting) and change-the-structure items (Kelly wants to be a singer. She wants to sing) where children are asked to repeat the same change in other similar sentences. These items are generally concerned with vocabulary or grammar. Although these tasks can be said to involve some use of language, they do not involve children in language use as we have defined it above, that is, in purpose- ful, creative and spontaneous language use in a particular situation.
106
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 107
Taped input: After giving an example of what is required, the instructions are ‘Now you listen and draw lines’
One Male Who’s the baby playing on the beach? Female That’s Sally. M Oh Right! Her Dad’s with her. F Yes. He’s putting the lunch out. Two M Look at the old man. F That’s Fred. He’s sleeping in the chair. M He read his book and then went to sleep, I think. F Yes.
Answer sheet:
Figure 4.6 Example of a listening assessment task (University of Cambridge ESOL Examinations, 2002, p. 23).
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
They are low in if not devoid of contextual support, and lack authenticity. Purpura (2004) discusses the place of these kinds of items in the assess- ment of grammar and vocabulary; he recognizes some value for them in certain assessment purposes (e.g., where the aim is to emphasize indi- vidual grammatical forms, such as in form-focused instruction) but repeats an earlier comment:
context-independent, discrete-point tasks, or those that lack authen- ticity of topic, are perceived by current and past students, teachers, administrators and content teachers as being ‘old-fashioned’ and ‘out-of-touch’ with their language learning goals.
(Purpura, 2004, p. 253)
Language-item oriented selected-response and limited-response items can be used with care by teachers and assessors:
• in the classroom, to diagnose specific underlying skills and abilities (e.g., ability to discriminate different sounds; knowledge of selected vocabu- lary and grammar items; ability to read or hear specific information in a text and select the most appropriate response from a given list);
• in external tests, to test specific contributing knowledge and skills. They are combined with language use tasks (extended-response tasks) to gain an understanding of children’s language use ability in the domains of language use being tested.
The use of selected-response and limited-response items therefore requires some careful decision-making; about their purpose (is it quite clear whether they are being used to assess language use, or to assess a language item?); about their role in the total assessment procedure (are there accompanying language use tasks in the procedure that assess lan- guage use ability?); and about their appropriateness for the learner group (are they cognitively appropriate for the young learners in question?). The use of the framework for the selection of tasks presented later in this chapter will help teachers and assessors to make decisions about the use of these types of tasks.
Selecting appropriate assessment tasks and procedures for young learners
What principles lie behind the selection of assessment tasks? Are some assessment tasks better than others?
108
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Language assessment tasks may be selected by the classroom language teacher, by the textbook writer or by others, for example other teachers in the school, test developers in the education department and in commer- cial testing companies. These language assessment tasks may stand alone, or may be part of an assessment procedure consisting of assessment across a number of tasks. For example, teacher observation, portfolios and self- assessment (described in the next chapter) are assessment procedures.
The following are considerations in the selection of language assess- ment tasks and procedures.
Some first-base principles to guide the selection of assessment tasks and procedures
The following are some first-base guiding principles on selecting tasks and procedures for the assessment of young learners. These first princi- ples come primarily from young learner education and assessment. They are followed by more specific principles and analytic frameworks taken from the field of assessment.
• Select tasks and procedures to suit the characteristics of young learners: The characteristics of the learners being assessed – their age, their interests and motivations, their social and personal char- acteristics will be known by the teacher and assessors, who will then be able to select tasks and procedures to suit these characteris- tics. We have looked in depth at the characteristics of children at different ages in Chapter 1, at their approach to language learning in Chapter 2, and of the implications for assessment of each. Teachers and assessors need to take account of many factors based on their knowledge of the purpose for the task and the characteristics of the learning situation.
• Assess the learners’ most relevant abilities for language use: Teachers need to make sure that they assess those abilities that chil- dren need in order to be successful in their language learning. The cur- riculum usually establishes the range of knowledge, skills and abilities that need to be taught, and therefore need to be assessed. A commu- nicative curriculum would establish the ability to use the target language as a central goal, with other related goals, for example socio- cultural knowledge, learning-how-to-learn skills and language aware- ness included.
Assessing language use through tasks 109
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
If no curriculum exists then teachers need to make their own judgment about the relevant abilities to assess. Teachers can carry out a needs analysis, a survey of the kind of language abilities children need, at school, in the community, in near-future language use activities (such as a class excursion to a target language speaking environment). Theoretical frameworks of language ability like the Bachman and Palmer (1996) framework of communicative language ability and the Common European Framework (Council of Europe, 2001) can inform teachers about the elements of language proficiency that need to be assessed.
• Make assessment choices that ensure that assessment is valid and reliable and has a positive impact: Many questions need to be asked about assessment tasks that are used, both in the classroom and in external tests. Is the task suitable for all children? Is it assessing what it sets out to assess? Is the scoring of the task appropriate? Will the task have a positive impact, for example on learning, and on the children’s future progress? Ways to analyse tasks and procedures in relation to these important questions are outlined in the next section.
• ‘Bias for best’ but maintain high expectations: The best assessment tasks and procedures are those that give children the chance to show their best performance. Swain (1985) coined the term ‘bias for best’ to convey this idea. We need to do everything possible to give children the chance to do their best. If a child’s performance is not strong, is that because there were factors in the task or the procedure preventing the child from showing what he or she could really do? Did the child have adequate time? Were the instructions given in a way that could be understood? Were there background noises – children playing outside – causing him or her to lose concentration? Were there cultur- ally based references in the task that the child had no experience of? Was the task or procedure sufficiently motivating for this particular child? The analytical tool in the following section provides many more questions that can help teachers and assessors to give children their best chance.
If teachers ‘bias for best’ there is no implication that the task should then hold low expectations of children. Children will often rise to expectations, if they are given the right conditions and support, and benefit from experi- encing success in challenging tasks. Sometimes additional support can be given within an assessment task to assist those who, with just that little extra help, can achieve the task and experience success.
110
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
• Engage learners intellectually: Assessment tasks can be appropriate in terms of hitting the right developmental and proficiency level for the children concerned, but at the same time they may be relatively lacking in intellectual challenge for the children. Sometimes a simple task is required and can be balanced with more intellectually challenging tasks. The issue here is asking teachers to question their choice of tasks overall – is there a degree of intellectual challenge for the children within at least some of the tasks?
Children might like the following task because they can do it – it is easy to colour in, and easy to hear the same sentence patterns repeated with substituted words.
Instructions: Colour in the picture of the house and garden with the right
colour when you hear the sentences.
Spoken sentences: The house is yellow.
The trees are green.
The car is red.
Etc.
This task gives the teacher information about whether children know their colours, but little else. Making it more intellectually challenging might mean bringing in a story or a problem. The following is a very simple example of how to make a listening task more interesting for young learners. Children need to know a small amount of additional vocabulary, but it would be possible to encourage them to listen for the meaning in the stream of language they hear (that is, they don’t have to understand every word to be able to carry out the task).
Instructions: Mimi is looking for some things in her house. Can you help her
find them? If you find what she is looking for, colour it in in the right colour!
Her book is red. Can you help her find it?
She has lost her blue toy elephant. Is it there?
Engaging learners intellectually is easier as they get older and as their proficiency grows. General knowledge and curriculum-related topics (from science, social studies, etc.) can be included to enhance interest and motivation.
• Draw from multiple sources of information: It is important to draw from many sources of information when decisions are to be made
Assessing language use through tasks 111
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
about children’s abilities, especially in high-stakes situations. Teachers should, whenever possible, collect data from a number of different tasks, selected to observe the desired range of behaviour. They should use different procedures if they can – observation, portfolios, self- assessment, quizzes and tests to make sure they gain the most accurate and composite picture of the child’s abilities. Making a decision about a child’s performance based on one source of information is ‘danger- ous and maybe even foolish’ (Brown and Hudson, 1998). External testers are limited to collecting information from one test – perhaps with around six tasks in it. Because of this the tests are carefully designed and thoroughly trialled. Despite this, external tests are not able to capture the breadth and depth of the child’s knowledge and ability. They are used for particular purposes (providing information about progress in comparison with others; giving information about progress on specified areas of the curriculum for motivational pur- poses) and simply cannot collect information about the full scope of the child’s abilities in the same way as the classroom teacher, who has multiple sources of information available to her during the course of her teaching and classroom-based assessment.
These principles are first-base considerations for the selection of assessment tasks and procedures. We will now turn to further ways of analysing tasks and procedures to ensure that they are the best for the children in question, for the assessment purpose and the assessment situation.
Frameworks to analyse assessment tasks and procedures systematically for selection
The assessment field provides us with frameworks and tools to look more closely at the tasks and procedures we use in assessment to make sure they are valid and fair, or ‘useful’. The rest of this chapter describes com- ponents of an assessment framework that can be used as an additional tool by teachers and assessors to check their selection and design of assessment tasks and procedures. The framework might be understood best through discussions in professional development groups, or in higher degree courses. The framework can be used to enhance teachers’ professional understanding, to raise awareness of issues in task selection and design, and to guide opportunities for close planning and analysis of assessment tasks and procedures.
112
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
The concept of ‘usefulness’: ensuring that assessment tasks and procedures provide ‘useful’ evidence
The concept of ‘usefulness’ (Bachman and Palmer, 1996) incorporates the idea of fairness, that is, the idea of ensuring that each child receives a score that most closely represents his or her abilities. ‘Usefulness’ includes important ideas that help us to ensure that assessment proced- ures will give assessors the best evidence for assessment decisions about children’s language use ability. The framework of ‘usefulness’ is a first-base approach to assessment used in formal assessment situations. There is some debate in the literature about the nature and role of valid- ity and reliability in formative classroom assessment. This issue will be addressed below, and in Chapter 5. Bachman and Palmer suggest that the following questions should be asked about assessment tasks and procedures:
• To what extent are results reliable? ( To what extent would the child get the same results if another teacher or assessor were to assess their work, or if they were to assess it in the same way again the next day?)
• To what extent is there construct validity? ( To what extent are the inter- pretations that teachers and assessors make on the basis of an assess- ment meaningful and appropriate?)
• To what extent is the assessment task authentic? ( To what extent does the assessment task reflect the kind of language children use in the classroom, or need in situations outside the classroom?)
• To what extent is the assessment task interactive? ( To what extent is the child’s language ability involved in accomplishing the task?)
• To what extent is the assessment practical? (Are there sufficient resources for the task to work in the assessment situation, and with young learners?)
• To what extent is the impact of the assessment positive? On the learn- ers? On ourselves as teachers? On parents? On society?
Each question is dealing with a particular quality of ‘usefulness’, summar- ized below. It is not possible to say that a procedure is absolute, that is, that it is ‘reliable’ or ‘authentic’ but rather the quality is described in terms of degree. This assessment procedure has high degrees of reliability; this procedure is highly interactive.
Assessing language use through tasks 113
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
To what extent are results reliable? Reliability refers to the consistency of the scores that teachers and assessors give learners – it is a measure of the degree to which an assessment procedure gives consistent results. We want to know that the results students are given are the same results that we, or another assessor, would give them for the same activity or group of activities on another occasion. If a teacher assesses a child’s work at a higher level of ability than another teacher, then there is a low degree of reliability in the scoring of this assessment task. If we give a student different assessment tasks to assess the same ability, but have different results, this is also cause for concern about the reliability of the score.
To what extent is there construct validity? Construct validity pertains to the meaningfulness and appropriateness of the inferences about students’ ability and the decisions teachers and assessors make about students on the basis of the assessment procedures used. If, in the assessment of chil- dren’s language ability, it is decided that we are aiming for their ability to use language, for example to tell the class something about their weekend, or to read and comprehend a short paragraph, that is our ‘construct’ or the the- oretical definition of the ability we are looking for. Teachers and assessors therefore need to be sure that the interpretations of the scores on an assess- ment task are a reflection of the nominated construct and very little else (Bachman and Palmer, 1996). If the aim is to assess the ability to ‘engage in a simple conversation’ it is important to make sure that the conversation is not concerned with concepts that are too difficult for the child (for example, concerned with things in the future that the child is unable to imagine yet). The child may be able to talk about home and school, or about objects and pictures placed in front of him or her, but being asked to talk about abstract concepts hampers the child’s ability to engage in the conversation success- fully. The inferences that would be made about the child’s ability to engage in a simple conversation about such a topic would not be valid.
To what extent is the assessment task authentic? Teachers would gen- erally want assessment tasks to be as authentic as possible, that is, they want children to be using language in tasks that are relevant and natural to the child’s world. Paper-and-pencil tests have little if any authenticity to the child’s world; writing e-mails to pen-pals has some authenticity for upper elementary foreign language learners keen to find out about chil- dren their own age in the target country; playing a fun game in the foreign language has high authenticity for young learners; and listening to a teacher’s talk on social studies has very high authenticity for a second
114
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
language learner who is studying in a mainstream classroom. Because young learners are growing, and changing in their use of language as they develop cognitively and socially, there is a question concerning the lan- guage they should be learning (and being assessed for). Should we be looking beyond the current world of the child to the future world? Cameron (2001) suggests children can be helped to grow into the lan- guage they will need as they grow older.
What ‘real language use’ (Skehan 1995: 23) is for these children is not obvious; it might be seen as the language used by native speaker 7 and 9 year olds, but by the time they have learnt it, they will be 9 and 10 years old, and will no longer need to talk about, say, teddy bears or dolls. The best we can do is aim for dynamic congruence: choosing activities and content that are appropriate for the children’s age and sociocultural experience, and language that will grow with the chil- dren, in that, although some vocabulary will no longer be needed, most of the language will provide a useful base for more grown-up purposes. (Cameron, 2001, p. 30)
To what extent is the assessment task interactive? An interactive task engages children in using the language knowledge and skills that are being assessed. Without interactiveness there is no evidence of language use on which inferences can be made about the child’s proficiency in the language. To give a straightforward example, if teachers give children a task that asks them to sort objects into different colours without having to listen to any instructions, or without having to say anything (that is, without having to use language in any way) the task is not interactive.
To what extent is the assessment task practical? An assessment proced- ure is practical if the necessary resources are available for it to be imple- mented. If too many resources (time, space, energy, materials) are needed, then the procedure will be impractical. A teacher might decide to assess all of his 45 foreign language students in a ten-minute individual oral interview. But this will take over seven hours. The ten-minute interviews are impractical in his teaching situation, with the present resources he has available. It might be practical for someone else’s teaching situation. Practicality is as important as the other qualities of usefulness: practical- ity is needed if assessment procedures are to be ‘useful’.
To what extent is the impact of the assessment positive? Impact is the effect of the test on teaching and learning, and on the many people involved
Assessing language use through tasks 115
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
in the assessment process – the children, their parents, their teachers, their school community and the community at large. Impact may be positive or negative. Teachers and assessors would, clearly, seek positive impact for their assessment procedure. Impact on teaching and learning is also known as washback. Assessment will not automatically have a washback effect on teaching and learning (Alderson and Wall, 1993), though many teachers are aware of signs of what appears to be positive washback (e.g., motivation to learn) and negative washback (discouragement of students, narrowing of the curriculum) in their classrooms. Positive and negative impact on teach- ers and on the community can be seen from high-stakes external tests, as I discuss in Chapter 9. Teachers should aim for high degrees of positive impact to help to ensure that the procedure is ‘useful’.
A ‘useful’ assessment procedure is therefore one that (to varying degrees) is characterized by the presence of all of these qualities, each prioritized to be as high as possible but in balance with the whole. Teachers and assessors can use the checklist provided to evaluate their assessment procedures. Evaluation can be done most effectively in group discussions in which teachers come together to talk about each of the qualities. Together teachers and assessors can consider each quality, and ensure that each is prioritized as much as possible according to the purpose of the assessment procedure, and the teaching situation (the nature of the children, the resources available and so on).
Recent thinking about validity, reliability and impact in classroom assessment
In recent years researchers in generalist education (e.g., Black and Wiliam, 1998; McMillan, 2003; Smith, 2003) have considered the nature of forma- tive classroom assessment (for a definition of formative assessment, see Chapter 5), and whether and how the concepts of validity and reliability apply to this type of assessment. Formative classroom assessment is carried out for the purpose of improving learning and can be embedded in instruction in an iterative and complex way, especially in what we have called on-the-run assessment. It has been suggested that there should be different ways of looking at validity, reliability and impact in formative assessment. Writers suggest, for example, that there is validity in forma- tive classroom assessment when an assessment decision has resulted in more student engagement, and when progress can be identified: ‘Validity is a characteristic that refers to the soundness, trustworthiness,
116
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
or legitimacy of [teachers’] assessment decisions’ (McMillan, 2003, p. 9). They suggest that reliability should be de-emphasised in formative assess- ment, since teachers are typically interested in how well the student does in the task, rather than finding out how well the student has performed in relation to other students (Smith, 2003, p. 5). They suggest that reliability can be checked in classroom formative assessment, through the collec- tion of sufficient observation data over many tasks. They also argue that the impact of classroom assessment can be evaluated through a consid- eration of the intended and unintended consequences of teachers’ deci- sions, for example: Do students acquire deep understanding as a result of preparing for an essay test? Do students believe they are capable of learn- ing new knowledge after self-monitoring of practice exercises? Does the class demonstrate needed skills or is remediation needed? (McMillan, 2003, p. 9). These new ways of looking at validity, reliability and impact in classroom assessment extend the constructs outlined earlier in the chapter (though some say, qualitatively change them) and contextualize them in the realities of formative assessment in the classroom. Research and thinking in this area are relatively new, especially in language assess- ment, and more research is needed to understand the relationships and differences between the constructs in formal and formative assessment. Meanwhile many would advocate that teachers keep a close eye on the characteristics of ‘usefulness’ as they go about their formative assessment, and at the same time accept that some yet-to-be-fully-understood adjust- ments are likely to be needed to accommodate the realities of formative assessment.
A framework of task characteristics: a framework to analyse individual tasks
Having considered some broader guiding questions about assessment tasks and procedures, we can now look at specific questions about individual tasks and check their characteristics. Is the task authentic, that is, does it reflect a real-life task that children are preparing for? Is it a suit- able and fair task for all children? If the task is suitable for most children, are there some children who might need additional support if we use this task?
The following framework (presented in full in a checklist in Table 4.4; see Appendix to this chapter) guides teachers through a close analysis of the characteristics of a task, so that they can check exactly what is
Assessing language use through tasks 117
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
happening in the task and how it might influence the learners’ perfor- mance. Children are vulnerable to task differences; there are cognitive, language, interactional, metalinguistic, involvement and physical char- acteristics of tasks that can influence the difficulty of a task (Cameron, 2000). In my experience, assessment tasks and procedures for young learners are rarely analysed to check for characteristics that may influ- ence (often disadvantage) performance. Guides for young learner assess- ment tend to focus more on ways to assess, criteria for assessment and ways to record observations, without asking for more detailed analysis of the actual demands on children within those tasks.
Analysing tasks in this way takes time. The degree to which time is spent analysing an assessment task will depend on a number of situational factors. If the assessment is high-stakes, this type of analysis is essential. If a summative assessment is required with a clear reading of each child’s abil- ities, then it is worth taking time to apply the framework to the assessment tasks during the task selection process. If a group of teachers is learning how to apply the framework, and are working together to reach an agreement on the selection of assessment tasks, this will initially take time. Once teachers have internalized the framework through professional development activi- ties, they will be able to apply it to many of their classroom assessment tasks.
The categories of task characteristics are as follows:
Characteristics of the setting: The physical characteristics of the setting, for example the noise level and the seating conditions, are a com- ponent of the characteristics of the setting. Some tasks may be outside in the community – what are the physical characteristics in such cases? The participants in the assessment task (for example, the interviewer in an oral interview, or other children in a group task) are part of the charac- teristics of the setting, as is the time of the assessment.
Characteristics of the assessment task procedures (rubrics): The instructions that are used in an assessment task are a component of the characteristics of assessment task procedures. The way that instructions are given (for example, by the teacher, in writing, on a tape), the length and difficulty of the task are some variables that characterize the task. The structure of the task is another element in this category of characteristics, for example the number and length of parts in the task, the sequencing of parts, and time allotted to each task. Methods of judging performance – the criteria for correctness and the nature of the procedures for scoring are included in this set of characteristics. To cater for young learners, questions
118
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 119
about the familiarity of the task structure, and the cognitive and social demands of the task are added to the original framework in this section.
Characteristics of the input: The input is the material contained in an assessment task that learners are expected to process in some way and to which they are expected to respond. In order to analyse the input in an assessment task, teachers and assessors need to ask questions about the format of the input and the language of the input. Questions about the language of the input are based on the Bachman and Palmer framework of language use described in Chapter 3 and relate to the areas of language knowledge described in this framework.
Characteristics of the expected response: The expected response is the response we are trying to elicit from children. We are trying to elicit a response through the instructions we have given, the task we have designed and the input we have provided. The actual response may be different from the expected response, as learners may respond differently because they have misunderstood something in the instructions, or may choose to do something in a different way.
The characteristics of the expected response may be described in terms of format, type of response, and degree of speededness. The format of the response may be analysed, for example, in terms of whether it is aural or visual, the expected language (first or target language) and its length. Type of response is concerned with the form of the response (does it involve choosing from among two or more responses provided? Is it an extended response?). We tend not to use degree of speededness in young learner language assessment; a speeded task is one where the score on the task depends primarily upon how quickly the test taker responds.
The language of the expected response, whether limited or extended, can be analysed in the same way as the language of the input, as above. This includes language characteristics and topical knowledge (compon- ents of language ability taken from Bachman and Palmer’s framework of language use).
Relationship between input and response: In some tasks, the input from the adult will be determined or adapted by the nature of the response from the child. This happens in oral interviews, when the adult hears the response and responds accordingly, for example, acknowledg- ing the response, adapting the content or level of difficulty of what he says
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
120
next. There is no relationship between the input and response in extended listening tasks (for example a story on a tape), or in reading or written tasks – these tasks do not adapt to the child’s response. A new type of reading or written task does, however, have opportunities for a rela- tionship between input and response to exist. Adaptive computer tasks can select the next question or set of questions (easier or more difficult questions) based on the response of the child.
Under this task characteristic, there are questions about the scope of processing – the amount of input that needs to be processed before the child can respond, and about the directness of processing – the degree to which the child is expected to process the response primarily based on the information in the input (or will he be required to draw on further information from the context or from his own topical knowledge?).
In the checklist in Table 4.4 (see Appendix), these characteristics are pre- sented as questions that teachers can ask as they analyse tasks. Since all questions will not be relevant for every task, teachers and assessors should select the questions that are relevant to the task in question. The questions should be checked against the characteristics of the learners. Thus, for example, ‘How many parts or activities are there?’ or ‘How much time is allotted?, is checked with the question ‘Is this suitable for the young learners who are being assessed through this task?’
Following are three examples of ways in which the framework can be applied to check the characteristics of the task. The first application of the framework ( Task 1) shows how the framework helps teachers and asses- sors to check that a task is fair to all children involved in the task. The second application ( Task 2) helps teachers and assessors to check if add- itional support is needed by some of the children in the class. Support involves modifying the task in some way to improve a child’s chances to succeed; in assessment tasks support is noted and taken into account when the child’s performance is judged. The third example ( Task 3) is of a different order. It shows how the framework can be used to check whether the assessment task reflects a real-life task, that is, whether it is an authentic task. This kind of analysis is needed when assessment tasks are required to reflect the real world and are designed to check that chil- dren are able to use language successfully in an authentic task. The result of an analysis can bring surprising insights into an assessment task that on first consideration appears suitable.
In the examples different formats are used. A template for analysis is provided in the appendix.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Task 1: Using the framework of task characteristics to check fairness
The framework of task characteristics can be used to check that a planned assessment task will be fair for all the children being assessed, that is, that they will receive a score that represents their real abilities in the language, and that there is absence of bias.
Example 1: Description of assessment task 1
Planned assessment task: Children respond to teacher’s questions
about pictures painted by children and
pinned on the wall. ( This task is adapted
from Reilly and Ward (2000, p. 82).
Characteristics of learners: Year 1, beginning EFL learners (aged 5).
Learning context: The assessment task is embedded in the
teaching task in which they are drawing
pictures of their families and talking
about them.
This assessment task is embedded in the teaching task in Table 4.2 and is highlighted in bold.
Analysis of characteristics of assessment task 1
Characteristics of the setting: The setting is familiar; the children will be in their classroom. The children will be settled and their attention will be on the teacher. It is a whole-class activity with their teacher whom they know well.
Characteristics of the task procedure (or rubrics): The teacher will give brief oral instructions in the target language, modelling what is required, and will not be collecting information on children’s abilities until it is clear that the children understand what is required and are involved in the routine. A teacher aide will be available to assist.
There is only one section in the assessment activity: The time for the whole class activity will be ten minutes. Children will be asked either to volunteer (hands up) or to answer. They will be given time to respond.
Assessing language use through tasks 121
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
122
Table 4.2 Teaching task: Children draw pictures of their families and talk about them
Preparation Core activity Follow up
Language To learn/revise Identify orally Identify members learning learnt lexis of members of of family in photos goals (other family members. their family brought to school. curriculum Practise ‘This in paintings. goals not is . . .’ ‘Her/his specified here) name is . . .’
Understand and respond to ‘Who’s this?’
To listen to and predict meaning from context to other information about the family, supported by pictures/context.
Teaching Children are Children draw Children are activities shown photos of pictures of their encouraged to
teacher’s family, families. bring in and are told ‘This Children tell the photos of their is my sister. Her teacher families and talk name’s . . ...’, and individually about them. further information who they are in photos (with painting, and pointing, contextual about their support). pictures.
Assessment task In groups, children
listen and respond to teacher’s questions about the identity of the pictures on the wall. E.g. ‘Who’s this?’ ‘Whose picture is this?’ and ‘What colour hat is she wearing/is her hair/ are her trousers?’ asked in contextualized ways (pointing to pictures, stressing important words).
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
The teachers’ judgment of performance will be by observation, with notes made, but backed up with an audiotape-recording. Children are not aware that this is an assessment activity.
Characteristics of the input: The input will be spoken target language, contextualized with pictures painted by the childen and pinned on the wall. The content of the pictures is familiar, as children will have just had the painting session. The vocabulary and contextualized structures will have been introduced and reinforced in the preparation activities. The teacher’s language will be at the discourse level, but target vocabulary and structures will be stressed through her talk using contextualization strategies (relevant pointing, verbal stress, expressions, repetition). The topic of talk is personal, and within the conceptual capacity, interest and experience of the children.
Characteristics of the expected response: The children are familiar with the teacher–whole group questioning on pictures that all can see, and on raising hands/being selected to respond. Their expected response is that they will show that they understand the teacher’s contextualized ques- tions in the target language, and provide short spoken correct answers in the target language. ‘It’s/That’s Mima’s brother’, ‘That’s my sister’, ‘It’s green/red/blue’, ‘They are white’, etc.’ Sentence-level answers without cohesive devices are expected. ( The emphasis is on understanding as these are beginning learners.)
The topic of the response is personal, and within the conceptual capac- ity and experience of learners.
Relationship between input and response: Talk is reciprocal – the teacher encourages and waits for a response, and provides feedback and comments on children’s response. The topical knowledge needed is not all provided by the context of the activity i.e. children must rely on their own knowledge of who painted the pictures and what relationship they are to the painter. The task is appropriate to the cognitive, social and physical maturity of the children.
Findings of analysis of task 1: Is the picture-drawing task fair for all children?
The analysis of this assessment activity for young children reveals that even in such a seemingly simple assessment activity, where the teacher is
Assessing language use through tasks 123
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
evidently catering well to the physical, social, cognitive and language needs of young learners, two weak points in the planned task have emerged.
Firstly, this is a whole-group assessment activity and is 10 minutes long. The teacher is counteracting the possibility that only more confident children will volunteer, by also naming children and asking them to respond. However, it will be difficult for her to assess all children before their attention spans wane. She will need to find ways to ensure that all children are assessed. A practice effect will also emerge in a longer activity, that is, children will hear other children’s responses as the task progresses, and this will mean that she may not be assessing chil- dren’s understanding, but their memory of what others have said. This is a threat to the validity of the task she will not want if she is intent on checking each child’s understanding (rather than teaching). The teacher may split the children into two or three groups, with other groups doing independent activities so that she can check all children in one group at a time.
Secondly, and much more seriously, the teacher is asking the children to identify the people in the pictures, but many children will not know who painted the picture or what their relationship may be to the painter. This expectation makes the activity much more difficult for some chil- dren in the group. Some children might not know all the children’s names, or may not have been concentrating on who painted what. Some will have better memories than others. There may therefore be a potential threat to the validity of the task for some children because of unfamiliar- ity with the content of the task.
Experienced teachers would quickly redress this weakness in the assessment task by altering the task, building in ways to help the chil- dren to know who it is in the picture. For example, the teacher might plan to choose a limited number of pictures with a range of family mem- berships with which to ask questions so that children will know who painted each one and who is depicted. She might decide to assess in smaller groups, using the teacher aide to supervise the class while she does this.
This is a simple example of how an assessment task can be analysed to check that it will work fairly, in this case, so that all children will have an opportunity to participate and to demonstrate what they know and that the teacher will collect the best language sample from each child without over-supported performance.
124
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Task 2: Using the framework of task characteristics to check additional support needs
For our purposes, support involves the modification of a task involving changes to the nature of the planned task or task-in-action in order to improve each learner’s chances of success. Support is a complex issue in assessment but cannot be ignored in the assessment of young learners. As part of sound educational practice, teachers and assessors of young learners are generally committed to ensuring some kind of success for their learners. Therefore, where possible, they will include additional support during the task once they are aware that a child has reached his or her limit of ability or is distressed.
Additional support may be planned, so that the characteristics of the task are changed for the targeted children from the beginning. This is commonly called task differentiation. Task differentiation is a recom- mended assessment strategy in mainstream classroom assessment in England (SCAA booklet, UK) and is valuable when it is clear before the assessment takes place that some children will need support to be suc- cessful in the task. Another term for support strategies used in planned assessments, usually formal assessments, is accommodations (see Butler and Stevens, 1997; see also Chapter 9).
Additional support may also be given during the task; that is, the teacher adjusts the task once it is clear that the child is not successful. In a classroom assessment situation, the teacher may prepare the children as planned and set the the children to the task; he or she may then walk around and make sure, by observation, that children have understood the requirements of the task. The teacher can help those who did not under- stand what they had to do, making note of the fact that these individuals needed this additional help. In a role-play task, a child may need some prompting to get started.
In both types of situations, teachers and assessors must make profes- sional judgments about the child’s performance and take into account the additional support given in their judgment of the child’s ability. This is not always easy. Although support adjustments appear unpalatable to measurement experts, who aim to see exact comparisons amongst learners, this practice reflects the essence of the reality of young learner teaching and assessment. In classroom assessment, the teacher’s profes- sionalism is critical; he or she must be relied upon to make the final judg- ment, balancing demand and support to come to a final decision about the ability of the child.
Assessing language use through tasks 125
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
126
Providing additional support is also a teaching mechanism, and in teaching is aligned closely with strategies teachers use to scaffold learn- ing as we have discussed in Chapter 2. The need to continue to apply support in assessment tasks for young learners when it is needed by individual learners or groups of learners is perhaps a reason why teach- ing and assessment are not easily distinguishable in young learner assessment contexts (Rea-Dickins and Gardner, 2000; Teasdale and Leung, 2000).
Some opportunities for additional support for learners in assessment tasks might be as follows:
• Characteristics of the setting: e.g. Support can be given to a child who is unsettled and nervous by placing him in a quieter setting.
• Characteristics of the assessment task procedures (or rubrics): e.g. Children who will clearly have difficulties with the instructions for the task might be given the instructions in their first language. A familiar task is chosen, so that all children can be successful.
• Characteristics of the input: e.g. The text used might be shortened for those with lower proficiency.
• Characteristics of the expected response: e.g. A child may be given extra time for the expected response.
• Relationship between input and response: e.g. A teacher prompts a child with additional questions and suggestions to help him or her to complete the task.
Providing too much additional support, in the sense that this makes the task too easy for the child, will not help the teacher to make informed judgments about the child’s abilities, as we will discuss in Chapter 6. The degree of familiarity within each of the characteristics of the task will have an important influence on the type of support some or all children receive. Therefore, as researchers have observed (Nicholas, 1999), over- familiar, over-practised tasks may not challenge children to use language in purposeful and active ways, nor to extend themselves and show their true abilities in the tasks.
The following example of a task analysis is similar in many ways to Example 1 in that we are analysing the task to check for fairness; however, in this case we are checking for the suitability of the task for a particular learner (or learners), and then planning support needs.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Example 2: Description of assessment task 2
Planned assessment task: Middle elementary children write a report
about an Australian animal they have been
researching. They will be expected to write
the report independently, but the teacher will
be at hand.
Characteristics of learners: Year 4 ESL children in a mainstream class-
room in Australia.
Learning context: The assessment task takes place at the end of
a teaching cycle, where learners have been
revising their understanding about the
purpose of reports, how to organize the text
and content of report, and how to write
appropriate language. They have undertaken
a series of activities involving modelling of
reports, joint construction, reporting to an
audience and reflecting on processes.
Analysis of characteristics of assessment task 2
This analysis in Table 4.3 has been done jointly by a mainstream teacher and an ESL teacher, just as in more formal test situations groups of teachers and assessors might devise tasks together using the framework. The wording in Bold type indicates the areas of possible difficulty for the ESL learner.
Findings of analysis of task 2: Will the second language learner(s) need additional support in this assessment task?
The ESL learner has three areas of difficulty (see teacher’s comments in bold above). He is very aware of the fact that results will contribute to a final mark and is aware that he is likely to do badly. He knows that his language proficiency in English is weak, and his knowledge of Australian animals minimal. His nervousness about this and about his parents’ reac- tions to his marks means that he has difficulty concentrating.
The language of the input – the books that learners can refer to, is academically oriented, with difficult sentence structures, and nominal- izations (classification, habitat) and science words that are unfamiliar.
Assessing language use through tasks 127
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
128
Table 4.3 Analysis of characteristics of assessment task 2
Characteristics of the setting In the mainstream classroom. Independent Physical characteristics work. During morning language arts Participants sessions over a week. Time of task
Characteristics of the assessment The instructions are given orally by the activity procedures (or rubrics) teacher. She explains in detail what is
Instructions required, referring back to previous Structure modelling and joint construction sessions, Time allotment and showing models of completed Scoring method reports.
The teacher talks about a good piece of work (that is, outlines the criteria for success).
Marks will be given and contribute to the final mark for the year.
The ESL learner is aware of the assessment and is nervous about his parents’ reaction to his final rating.
Characteristics of the input The input is made up of reading texts from Format the library, found by the students themselves. Language of input Many have pictures to aid understanding; the
language is academic (social studies) and is targeted to elementary age (mother tongue) learners. The topical knowledge required is concerned with animals, their description, classification, habitats etc.; skills in using the library are needed, as is a cultural understanding of how to work independently (that is, without teacher-centred input).
For the ESL learner the academic language is difficult, and so is the topical knowledge – he knows little about Australian animals. He is also not familiar with library research, nor with working without teacher direction.
Characteristics of the A two-page report is required, following a set expected response format provided in models. (Classification
Format [ What is it?]; Description [ What attributes does Language of expected response it have?]; Place/Time [ Where is it? When is it?]; Topical characteristics Dynamics [ What does it do?]; Summarizing
Comment).
The report should exhibit the appropriate structure, and language features of a report
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 129
The report genre required is new to the ESL learner; this is not a writing task that he was familiar with in his previous school environment in his home country. He is unfamiliar with Australian animals, though he has read a little about them in his first language at home.
The analysis of the task highlights the ESL student’s likely weak areas in the assessment task. The mainstream teacher and the ESL teacher can decide together what support can be given to the student. The fact that this is a formal assessment situation means that they should take note of the additional support they give, in order that they can come to a final result and report that gives the ESL learner and the parents an understanding of his progress in relation to the rest of the class. (In some assessment situations, it might be possible to give the ESL learner an independent mark based on his own progress. This, of course, is more pedagogically appropriate.)
The teacher can therefore plan to give the student the following add- itional support:
• Reassure the learner that his or her parents will also receive a profile report outlining his progress in ESL terms as well as in comparison to mainstream learners.
• Select texts (with pictures, and accessible language) for the learner (that is, avoid the need for library skills at this point).
• Give the learner guiding questions to answer reflecting the required stages in a report genre.
Table 4.3 (continued)
genre, and contain relevant information, with pictures, about the animal chosen.
The level of difficulty of the concepts is suitable for Year 4 age learners.
For the ESL learner, the expectations for the response (genre, language, content) are very difficult.
Relationship between input and There is no reactivity. The amount of response information needed by students is mainly
Reactivity included in the books read, though most Scope of relationship children will rely on previous knowledge of Degree of relationship the animal to assist them to complete the (degree of contextualization) report.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Task 3: Using the framework of task characteristics to check that an assessment task reflects the real-life task and is therefore relatively authentic
We will now turn to the use of the framework for a purpose of a different order. There are some assessment situations where assessors need to check that the task is an authentic one, that is, that it reflects the charac- teristics of a real-life task. Classroom teachers of young learners will not often have to do this type of checking, as most assessment tasks in the classroom come straight from the kinds of tasks usually carried out in the classroom – they are not required to reflect tasks ‘in the real world’ as they are ‘authentic for the language learning classroom’. In contrast, teachers preparing second language learners for mainstream classrooms may sometimes need to check that a task reflects the characteristics of main- stream learning tasks (e.g., giving a presentation to the class about a social studies project; taking notes from a video on a science topic; inter- viewing a visitor). In a similar way, test designers in the United States are currently working on tests in which they will assess if children are devel- oping the academic English language skills they need for participation in mainstream classes. These designers need to select test tasks that reflect the authenticity of the mainstream classroom, and the application of a framework of task characteristics can help them to do this. (I will describe the process of development of such a test in Chapter 9.)
To use a reasonably simple example to illustrate the analysis of tasks for this purpose, we might envisage a situation where young learners are being prepared by the teacher to go out and interview tourists in the city. This type of activity has been proved successful in Hong Kong where a teacher, Zoe Leung, has taken new arrivals from mainland China in the beginning stages of English, to interview English-speaking tourists near the tourist spots of Hong Kong. In order to assess children’s ability to do this, the teacher sets up a highly comparable assessment task in the classroom before they go out and ‘pester’ tourists. The teacher is checking that her children are ready to attempt this activity, and that they are likely to expe- rience some success. Note that this example of task analysis is focused on authenticity, not, at this point, on checking fairness or support needs.
Example 3: Description of assessment task 3
Planned assessment task: Interviewing a tourist with prepared intro-
duction and questions with limited, mostly
predictable, responses ‘May I ask you some
130
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 131
questions?’ ‘Where do you come from?’ ‘How
long have you been in Hong Kong?’, ‘Where
are you staying?’
Characteristics of learners: 11-year-old beginning EFL learners in Hong
Kong. (Students recently arrived from main-
land China, with little experience talking to
non-Chinese speakers.)
Learning context: Students have been focusing on the lan-
guage requirements, and practising this task
for several weeks in their English classes.
The teacher has taught ways to be polite in
the situation, and prepared questions with
them.
Analysis of characteristics of assessment task 3
Characteristics of the setting : In the real-life task, the interview setting will be in a crowded tourist area in Hong Kong, and there may be difficulties hearing every word that is spoken. The learners will be nervous because they will not know if they will be rebuffed rudely, or whether they will be able to make themselves understood, or understand what is said to them. They will also be ‘on their own’ without teacher support, though they will know that the teacher will be hovering nearby (but not within earshot). The teacher will advise them to make a beeline for tourists who look Caucasian, because at a guess they will speak English, but they may have a range of different English accents – American, Australian or English, perhaps South African, and they may not even speak English at all! The characteristics of the real-life setting are therefore quite unpredictable for the learners. The task involves students talking to unknown adults; this will involve certain expectations that arise when young people talk to older strangers (politeness, readiness to with- draw, formal questioning only, etc.).
In the assessment task, the level of anxiety will be difficult to duplicate. It may be possible to ask English-speaking visitors to come to the class to be interviewed. The noise level can be raised in the busy classroom by asking two or three visitors to come to be interviewed simultaneously.
Characteristics of the task procedures (or rubrics): In the real-life task, the procedures for the task will be informal in the situation once the
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
approach has been made. However, the teacher will need to give careful instructions on how to approach the tourists, on the kinds of questions to ask (if the tourist is willing to participate) and on the way to thank the tourists and withdraw. The judgment of performance, even in the real-life task, will be through teacher judgment and student self-assessment of the success of the interaction. The teacher will talk to the students about and make formal the criteria for successful interviews, including politeness on the part of the students, use of English in a way that is understood, under- standing and recording of the tourists’ responses (if they were willing to take part), and polite withdrawal. Thus, to some extent, this real-life task has characteristics of a classroom teaching task – it is structured carefully by the teacher, will be assessed by the teacher and will be supported by the teacher if needed.
In the assessment task, the same procedures will be used. In this respect the real-life task and the classroom assessment task are comparable.
Characteristics of the input: In the real-life task, the language that the tourists use is somewhat unpredictable at the commencement of the interview, but at least structured by the questions the students will be asking. Thus once introductions have been made, and consent given, a question like Where do you come from? is likely to elicit a response I come from Australia. Accents may be unexpected, and angry, impolite or non-comprehending responses to Do you mind if I interview you? may happen. The language of the input and the topical characteristics will most likely be limited by the students’ questions, and by the tourists recognizing the youth and beginning learner status of the students. It would seem that the real-life task is challenging, but related to the cognitive, social and emotional maturity of the students if the teacher is present for support.
In the classroom assessment task, the native-speaking visitors would need to be primed to be somewhat unpredictable at the beginning of each student’s or group’s interview with them. The questions and answers would probably match the real-life task. The reality of an unsolicited approach and introduction – Excuse me, may I interview you for a school project? – would, however, be difficult to reproduce in the classroom.
Characteristics of the expected response: In both the real-life task and the classroom assessment task, the expected response from students is the same, although they would have to deal with more unpredictability and anxiety in the real-life task. In terms of language use, they are
132
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
required to ask prepared questions and to understand and record the tourists’ or the visitors’ responses to report back to the teacher. The teacher is close by but not within earshot, so conveying the responses to the teacher would be real meaning exchange in both situations. There may be some cultural aspects (topical characteristics) of the interaction in both situations that would need to be dealt with.
Relationship between input and response: In both situations, there is some potential for reactivity. At the beginning of the real-life task interaction will be more reciprocal, when the tourist may want to know more, or say something unexpected. Once the interview is underway, interaction will be less reciprocal as the interview follows the ‘script’ determined by the pre-planned interview questions. The scope of pro- cessing required from the learner is, once the interview is underway, only narrow. The processing of the response by the student will be supported by gestures (kind smiles, unfriendly frowns) and by other gestures, and perhaps repetition, so there will be a degree of contextualization to support the learner.
In the planned classroom task, the task will probably be less reciprocal at the beginning of the interview, since the visitor will already know why he has been invited. However, the teacher could take the step of priming the visitor to act as an unsuspecting tourist when he comes into the class, which will help to mirror the characteristics of the real-life task in the classroom.
Findings of analysis of task 3: Does the assessment task reflect the real-life interview task?
This is an example of how to analyse a task to check whether the assess- ment task reflects the characteristics of the real-life task. If English- speaking visitors were not able to attend the class (which is quite likely) then the task would need to be reanalysed using Chinese-speaking teachers or parents probably familiar to the students. Several more mismatches would then be likely to occur between the assessment task and the real-life task (the Chinese-speaking teachers would not be so intimidating; the children would know that they could resort to Cantonese, etc.)
Readers will hopefully have realized by reading this that analysing the task would help the teacher to clarify several of the task’s characteristics,
Assessing language use through tasks 133
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
134
and improve the planning process for both the assessment task (check- ing if the students are ready) and teaching and learning (checking on further teaching requirements to prepare the students for the real-life task). Since the support for these students is high, the teachers will need to make complex decisions about the students’ final marks for the task, as the task is now differentiated – the characteristics of the input and the expected response are different for the ESL learner compared to the mainstream students.
Since all classrooms around the world, whether foreign language or minority language learning classrooms, are made up of children with a range of abilities, backgrounds and needs, these types of decisions are regularly made by teachers in classroom-based assessment situations. Professional judgments balancing the nature of the task and the nature of support are at the heart of young learner assessment. The framework of task characteristics is therefore a tool that can help teachers and asses- sors to understand the way an assessment task will influence learners’ performance.
Summary
Task-based assessment is a kind of performance-based assessment. Performance-based assessment involves the observation of behaviour in the real world, or of simulated behaviour in a real-life task. The principles of alternative assessment, authentic assessment, criterion-referenced assessment and divergent assessment reflect, with variations, the basic idea that assessment is best done through samples of learners’ real lan- guage use, rather than through discrete-point items that assess aspects of the learners’ knowledge of the language.
Tasks are defined as involving learners in purposeful, goal-oriented language use, specific to a certain situation. Many classroom activities can be used as assessment tasks. Selecting tasks and procedures for assessment involves great care, as bad decisions can cause disadvantage for all learners, or for a particular group of learners, or for an individual. Tasks and procedures should, for example, suit the characteristics of the learners, bias for best, engage the learners intellectually, assess the most relevant abilities and draw from multiple sources of information.
If teachers have an opportunity to analyse tasks more closely, prefer- ably with other teachers, they will gradually gain a professional frame- work that will help them to analyse assessment tasks more efficiently.
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessment tasks and procedures need to be ‘useful’ (Bachman and Palmer, 1966). Tasks and procedures should be reliable (the learner should get the same results if another teacher were to assess their work, or if they were to be assessed in the same way again tomorrow); have construct validity (the interpretations that are made should be mean- ingful and appropriate), be authentic (the task should reflect children’s real language use), be interactive (they should involve language ability in accomplishing the task); be practical, and have positive impact (a posi- tive effect on the learners, teachers, parents and others affected by the assessment).
The analysis of tasks for ‘usefulness’ can be carried out using a frame- work of task characteristics (adapted from Bachman and Palmer 1966). Three purposes can be achieved using the framework – firstly, analysing to check if the chosen task reflects the intended real-life task, that is, whether it is authentic; secondly, to check fairness; and thirdly, to check for additional support needs amongst the learners. Examples of each of these types of analyses are provided.
The following chapter explores classroom-based assessment from the theoretical standpoint of this chapter – that the most effective assess- ment of language use is through performance in language tasks.
Assessing language use through tasks 135
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Appendix
Table 4.4 Template for checking task characteristics for young learners (adapted from Bachman and Palmer, 1996, pp. 49–50)
Task characteristics Yes/No and comments
Description of the learner group:
Task title:
Characteristics of the setting
Physical characteristics Are there distractions in the environment? Is the setting familiar or unfamiliar?
Participants How many participants are there? Who are the other participants? Will the participants intimidate the child or cause a less
successful performance?
Time of task Is the time of the assessment likely to influence the child’s
performance? (e.g., just before home time)
Characteristics of the assessment task procedures (rubrics)
Instructions Are the instructions in the child’s native language or the
target language? Are they aural or visual? Is the language at an appropriate level? Are the instructions conceptually appropriate for the age group? Is visual support given? Are examples provided? Other?
Structure How many parts or activities are there? How long are the instructions? Are the different parts of the procedure clearly distinguished from
one another? Is there a fixed or variable sequence in the procedure?
136
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Table 4.4 (continued).
Do the parts of the procedure differ in importance? How many activities or items are there in the procedure? Is the task structure familiar (Do children know what is expected)? Are the cognitive and social demands of the task at the level
appropriate to the age of the children? What are the demands on the child’s attention span, on the
length of time he or she has to sit still? Is there any variation of physical requirement to ease demands?
Other?
Time allotment How much time is allotted? Will the activity be limited in time so
that not all learners are expected to complete it or will it be long enough for all to attempt every task?
Other?
Scoring method How will the task be judged as correct? Will an objective score be used (i.e., will there be a single correct
answer)? Or will rating scales be used? Are all responses marked by the same teacher or assessor? If different teachers or assessors are used, are they involved in
checking their decisions together so that the scores are comparable?
Will the children know what the criteria are and how the scores will be given? How will this be done?
Other?
Characteristics of the input
Format
• How is the input presented? In language form (written or spoken)? Or in non-language form (aural or visual including gestures, pictures, graphs, etc.)? Or both?
• Is the learner familiar with the form in which the activity is presented?
• To what extent are the format of the task and the language of the input appropriate to the cognitive and social maturity level of the child?
• What are the demands on the child’s attention span, on the length of time he or she has to sit still? Is there any variation of physical requirements to ease demand?
• Will the format and input raise the interest of the child and encourage the child to participate in the task?
• Is the language of the input the learner’s native language, the target language or both?
Assessing language use through tasks 137
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Table 4.4 (continued).
• Have the vocabulary, structures and genres been taught and practised in different contexts? Have they been taught or revised recently?
• Is the input made up of single words, phrases, sentences or extended discourse?
• Is the input an item (a chunk of language or non-language information) or a prompt (an instruction)?
• How fast does the child have to process the information in the input?
• Is the input delivered ‘live’ e.g. by the teacher in the classroom, or is it ‘reproduced’ e.g. via an audiotape or by both means?
• To what extent can children draw meaning from the context (from pictures, graphs, objects, the environment)?
• Other?
The language of the input
• To what extent is the language supported with teacher’s explanations, with pictures, charts and realia? To what degree are the teacher’s explanations stressing meaning through painting, stress, repetition, gestures and facial expressions?
• What is the nature of the organization characteristics of the language – the vocabulary, syntax, phonology (sound system) and graphology (writing system)?
• What is the nature of the pragmatic characteristics (what functions are being met, in the input? What dialect or language variety is being used? What register? (e.g., is the language formal or informal?)
• Is academic language being used? • Is the language natural? • Are there cultural references and figurative language in the input? • What is the topic being discussed (personal, cultural, academic)? • Other?
Characteristics of the expected response
Format
• Is the expected response aural, visual or both? • What are the demands on the child’s attention span, on the
length of time he or she has to sit still? Is there any variation of physical requirements to ease demands?
• How is the response to be produced – in language form (written or spoken), or in non-language form (aural or visual including gestures, pictures, graphs, etc.)? Or both? Are fine motor skills required (e.g., to write or draw)?
138
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core
Assessing language use through tasks 139
Table 4.4 (continued).
• Is the expected response to be in the child’s native language or his target language or both?
• To what extent is the child familiar with the type of response required (e.g., use of familiar actions and routines)? Has the child practised this test many times?
• Is the expected response to be made up of single words, phrases, sentences, or extended discourse?
• What are the demands on the child’s attention span? Are fine- motor skills required (to write or draw)?
• To what extent is the child familiar with the type of response required (e.g., use of familiar actions, familiar routines)?
• To what extent has relevant language been taught and learned? • Have they been taught or revised recently? • To what degree can the children support what they say and do
with reference to the context? • To what extent is the format of the task at the appropriate
level of cognitive and social maturity? • Other? Language of the expected response
• What is the nature of the organization characteristics of the language – the vocabulary, syntax, phonology (sound system) and graphology (writing system)?
• What is the nature of the pragmatic characteristics (what functions are being met in the expected response)?
• What dialect or language variety or register is to be used? • Is the language expected natural? • Are cultural references and figurative language expected? • What is the topic being discussed (personal, cultural, academic)? • Other?
Relationship between input and response
• Will the child be expected to give a one-off response (non-reciprocal), or will there be some processing of the input in relation to further information or feedback given (reciprocal)?
• Is the task adaptive (the task/the next question is adapted on the basis of the response the learner makes)?
• Does the child have to process a lot of input, or a limited amount of input? Will the child be expected to process the response based primarily on the information in the input, or will he or she be required to draw on further information from the context or from his or her own topical knowledge?
• Other?
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511733093.005 Downloaded from https://www.cambridge.org/core. Wayne State Univ Libraries, on 22 Sep 2020 at 13:12:21, subject to the Cambridge Core