Introduction
Assessment can be defined as the tools, techniques and procedures for deciding what learners can and cannot do as a result of instruction. There are many different types of assessment, as well as a wide range of approaches and instruments. The purpose of this document is to introduce teachers to the principles and practice of taskbased language assessment. While taskbased language teaching is becoming increasingly popular internationally, and, indeed, is a cornerstone of the new Chinese language syllabus, many teachers remain unsure about how to use the principles of taskbased learning to assessment.
Task-based language teaching: An introduction
Before considering the issue of assessment, it is important to clarify the concept of taskbased language teaching. In order to do this, consider the following sets of instructions.
Set A
Evaluate a set of statements by deciding whether they are “true” or“false”.
Study a written passage and correct the grammatical mistakes.
Put some questions and answers into the correct sequence to make a conversation.
Set B
Ask someone what time the next train leaves for the city.
Order a bowl of noodles.
Invite a classmate to the movies.
What is the difference between these two sets? While both involve using language, Set A involves using and manipulating language for its own sake. The items in Set A require learners to display their knowledge of language. Set B, on the other hand, requires learners to use their knowledge of language to obtain goods (e.g. noodles) and services (information about train departure times): the final outcome of using language is to go beyond language to obtain goods and services and to socialize. For this reason, we call them communicative tasks.
Principles of taskbased assessment
The best way to find out if someone can do the kinds of things in List B above, is not to ask them to do a true / false exercise or a multiple choice test, but to get them to try and do the tasks themselves. The first principle of taskbased assessment, therefore, is that learners should be involved in some kind of communicative performance that goes beyond the manipulation of language for its own sake.
A second important principle, and one that flows naturally from the first, is that teaching and assessment should be closely related. It makes no sense to have a taskbased syllabus on the one hand, and a traditional grammar or vocabulary based assessment procedure on the other. This does not mean that grammar and vocabulary should not be assessed. However, they should be assessed in contexts using tasks that are similar to those that were used to teach the grammar and vocabulary in the first place. The advantage here is that teaching tasks can be sued for assessment purposes, the key differences being the criteria used for judging learners performance, and how that performance is recorded and reported.
Features of task-based assessment
Taskbased assessment should do three things. It should elicit performance that is directly related to the communicative objectives of the course, it should provide learners with opportunities to show what they can do with the language, and it should describe the criteria by which learners’ performance will be judged. In this document, we look at practical ways of doing these three things.
Purposes for assessment
There are six main purposes for assessment. These are:
·to provide learners with a record of what they have achieved
·to provide feedback on how well they are progressing
·to provide information on what learners can and can't do so that language items can be recycled
·to provide information to the educational system as a whole
·to place learners in classes and learning groups
·to motivate learners to take responsibility for their own learning
The type of assessment test that is used, the criteria that are used to judge learner performance, and the means of providing feedback will all be determined by the overall purpose of the assessment.
Types of assessment
This section looks at different types of assessment, from direct to indirect, from normreferenced to criterionreferenced. It looks also at the distinction between the assessment of proficiency and achievement.
Indirect versus direct assessment
The two test items below illustrate the difference between direct and indirect assessment.
ITEM 1
|
Unscramble these questions.
Example: When do you get up?
1. get at up your six does mother thirty?
2. you time go do bed what do?
3. take morning a shower do in you the?
4. eat friend does breakfast your?
5. time to what school do go you? |
ITEM 2
| You need to change the time of your English lesson. You call your tutor, but he is not at home. Leave a message on his answering machine giving your name and telephone number. Tell him why you can't make your lesson, and suggest another time. |
Item 1 is an indirect test item, because it only gives us indirect information about the communicative ability of the learner. It will tell us whether the learner can form wh- and yes/no questions in the simple present tense, but it will not tell us whether he or she can use this grammatical knowledge to communicate.
Item 2 is a direct item. If a learner can carry out this task correctly, we can reasonably assume that the learners would be able to leave a similar answering machine message in the real world outside the classroom. It is therefore a taskbased assessment item.
A comprehensive test will contain both direct and indirect items. However, the major emphasis should be on direct, taskbased items. Regardless of whether direct or indirect items are used, the items themselves should reflect the content that the students have been studying.
Norm-referenced versus criterion-referenced testing
Traditional assessment is normreferenced. A norm-referenced test compares learners with each other, and the results will reveal how the students performed in comparison with each other (who came first, who came second, and who came last in the group.) Normreferenced tests are designed so that some learners will do really well, some will do reasonably well, and some will do poorly.
A criterionreferenced test evaluates the individual against the test item itself, asking how well that person performed. On a criterionreferenced test, all learners all learners can do well, or not so well.
Indirect tests lend themselves to normreferencing, while direct, task-based tests lend themselves to criterionreferencing.
Formative versus summative assessment
Formative assessment is ongoing, carried out during a course and is intended to provide feedback to learners on their strengths and weaknesses as they progress through a course of study. Summative assessment, on the other hand, typically occurs at the end of a course or module of work and is intended to provide information to learners and some outside sources (parents, the school etc.) on how well the learners have performed on the course as a whole.
How to develop assessment tasks.
In this section, I will provides a stepbystep guide to the developing an assessment task.
STEP 1: Study the teaching materials that the students have been using. It will not be possible to assess everything that the students have learned, particularly is the test is an end-of-book rather than end-of-unit test. It is therefore important to make a representative selection of items.
For example, the first three units of Go for it!Book 1 introduce the following functions and structures:
Functions:
Introduce yourself
Greet people
Ask for and give telephone numbers
Identify ownership
Introduce people
Identify people
Structures:
Present tense to be
What questions
Possessive adjectives: my, your, his, her
Demonstratives this, these, those, that
Subject pronouns: I, he, she, they
Yes/no questions and short answers
Plural nouns
It is not feasible to test all of these items (as well as other items such as core vocabulary and the four skills) so a selection will need to be made.
STEP 2: Collect, adapt or modify input data for each assessment item.
Most assessment tasks will require learners to listen to or read a text or process information from a visual source such as a picture or a chart. The best way to get these is to modify or adapt them from the textbook itself. For example, Unit 3 contains the following reading task.
|
Read the letter. Draw a picture of Emma's family.
Dear Maria,
Thanks for the great photo of your family. This is my family. These are my grandparents and this is my mother. This is my uncle and these are my friends. My friends are Peter and Alice.
Your pen friend,
Emma |
This can be readily turned into an assessment item as follows:
|
Read the letter. Draw a picture of Steve's family.
Dear Lee,
Thanks for the great photo of your family. This is my family. These are my two aunts and this is my father. This is my sister and my friend. My friend is Patrick.
Your pen friend,
Steve |
STEP 3: Draft the assessment procedure (what the learner is required to do).
At this step, it is necessary to set out the procedure, along with a set of instructions to the learners for what they are required to so. Again, the test item can be readily adapted from the textbook. Here is an example for assessing the listening and speaking objective of greeting people, introducing yourself, and exchanging telephone numbers.
You are meeting classmates at your new school. Greet another student and exchange personal information (family name, your name, telephone number).
STEP 4: Specify the criteria to be used in evaluating learner performance
This is a crucial step, because the criteria are what differentiate a teaching task from an assessment task.
The task from step three already has one build-in criterion-three pieces of information have to be exchanged. Other criteria might be:
·the learner's pronunciation must be comprehensible;
·the learner must form grammatically correct questions using the appropriate form of the verb to be as well as the possessive adjectives my, and your;
·the learner must demonstrate an ability to being and end an interaction appropriately.
STEP 5: Decide how the results are going to be reported, by whom to whom
How the results of the assessment are to be reported will depend on the purpose of the assessment. If the principal purpose is to provide learners with feedback, giving them insights into their strengths and weaknesses, then qualitative feedback, perhaps supported, in summary form, by a criterionchecklist (see below) will be most appropriate. If the assessment is meant to be summative (e.g. at the end of a course, a term, a year), some form of number or letter grade may be required.
Criteria for assessing learner performance
Performance criteria are a crucial part of an assessment procedure or item. These criteria need to be related to the important learning outcomes. In other words, they need to reflect those objectives that the task has been designed to assess. Criteria should address a number of aspects of performance when assessing a group oral presentation, for example, the criteria would include:
·linguistic accuracy
·fluency
·presentation style
·content
·collaboration with others
Criteria can be relatively simple or quite elaborate. The simpler the criteria, the easier they are to use. However, more elaborate criteria provide more detailed and complete information on the learner.
Means of providing feedback to learners
Feedback can take many shapes and forms from quantitative feedback in the form of marks and grades through to qualitative feedback from the teacher. The form it takes will depend much on the purpose of the feedback which, in turn, will depend on the purpose of the assessment. Consider the following criteria for assessing the task of Giving directions.
Test item:
| You are going to have a party on Sunday. Tell your partner how to get to your apartment for the event. |
Assessment criteria:
|
Content
Accuracy with which directions are provided
Communicative strategies
Ability to check speaker's comprehension
Pronunciation and fluency
Use of appropriate intonation and stress patterns
Grammar and structure
Accurate use of imperatives |
These criteria could be incorporated into an observation sheet such as to following for giving feedback to the learners. If a number or letter grade is required, each of the descriptors (needs improvement / satisfactory / well done) can be given a number (3, 2, 1) which will give a possible total score of 12 on the task.
|
NAME: CLASS: DATE:
TASK: Giving directions
Content
Accuracy with which directions are provided
(needs improvement / satisfactory / well done)
Communicative strategies
Ability to check speaker's comprehension
(needs improvement / satisfactory / well done)
Pronunciation and fluency
Use of appropriate intonation and stress patterns
(needs improvement / satisfactory / well done)
Grammar and structure
Accurate use of imperatives
(needs improvement / satisfactory / well done) |