If you have been working in a CLB context, you likely have experience with multi-level classes. You may have worked with classes that are designated across several benchmark levels (for example, CLB 1-4), or with classes including learners with benchmarks levels higher or lower than the designated CLB level in one or more language skills. In this chapter, we use the term multi-level to describe both classroom realities.
In this chapter we present steps you can take to adapt your assessment practice for more than one CLB level. We begin by introducing class profiles.
“I have so many different levels in my class – how do I meet all of their needs?”
“How can I assess learners at different levels in my group without using multiple tasks and tools?”
How do I Decide which Levels to Plan for in Each Skill?
Whatever your context, your classes will likely include learners at varying levels. Deciding which levels to plan for begins with identifying the distribution of learners’ levels within each skill. This will help you determine where learners cluster and also identify the degree of benchmark variation in each skill. A quick and easy way to do this is with a class profile chart like the one below, showing a profile for a CLB 3 class of 20. Instructors who have started using these class profiles have found them very informative.
Example: CLB 3 class of 20 learners, assessed for one level
Most of the learners in this classroom cluster at working towards CLB 3 in the four skills (as shown in green). If this was your classroom, and you planned to assess at only one level, you would be teaching and assessing at CLB 3 in all four language skills.
Example: CLB 3 class of 20 learners, assessed for two levels
In the same class, if you decided to develop assessment tasks at two levels, you would identify the adjacent level that has the second largest cluster of learners in each skill, as indicated in pink below. Based on this analysis, you would likely develop assessment tasks at CLB 3 and 4 for listening, speaking, and reading, and at CLB 2 and 3 for writing.
Even in the second version, however, outliers exist. Two students are working towards CLB 5 and one towards CLB 2 in listening. If you are assessing at two levels, your tasks will not be specifically geared towards CLB 5 or CLB 2 in listening. You might find opportunities to add challenge to some of the tasks, or perhaps provide some individual listening tasks that reflect CLB 5. Conversely, for the learner working towards CLB 2, you may need to provide additional instructor support or scaffolding on tasks so the learner can complete the task. Your comments would indicate that support had been provided, (e.g., “You were able to complete this task with instructor support”) while also providing action-oriented feedback (e.g., “Listen for the stress in numbers. 15 = fifTEEN, and 50 = FIFty”).
This action-oriented feedback will help the learner move forward, and will prevent the discouragement that he or she might experience faced with repeated “not yets” on the assessment criteria. Most importantly, however, these outliers in listening are being assessed on their reading and writing skills, which are within the levels adopted for assessment, according to the profile.
Similarly, there are outliers in reading – one learner has completed CLB 4 and two have completed CLB 1. Again, for the CLB 4 learner, you might provide a challenge beyond the CLB 3-4 tasks, and for the CLB 1 learners, significant scaffolding or fewer required responses.
In writing, the assessment tasks will be focused at CLB 2 and 3. The outliers have completed CLB 3, so are working towards CLB 4. In this case, you can assess and confirm their CLB 3 level, and you might be able to provide “higher-level” feedback beyond the identified criteria that will help the learner continue to progress. In this fashion, feedback will focus on learners’ current needs, with outlier students building skills.
If you want to develop assessment tasks across three benchmark levels, you could use the same process to develop a class profile and identify the levels to assess.
TO CREATE A CLASS PROFILE:
Enter the range of CLB levels in the left hand column.
Enter the total number of learners at each level for each skill.
Identify assessment levels:
Identify the CLB level in each skill with the largest cluster of learners (the core level you will assess).
Identify the adjacent CLB level in each skill with the second largest cluster of learners (the additional level you will assess).
Record the level(s) to assess in the bottom row.
How do I Modify my Assessment Tasks to Address more than one Level?
Adjusting assessment for more than one benchmark involves more than simply changing the scoring on your assessment task; it requires the careful adjustment of the task for each level assessed. If success on a CLB 6 reading task is a score of 24/30, for example, a score of 20/30 doesn’t necessarily mean that the learner is working at CLB 5. To gain an accurate assessment of CLB 5 work, the task must be carefully aligned to the expectations for that benchmark level.
You can modify an assessment task to address multiple levels by following two steps: first analyzing the task, and then modifying it to fit the levels being assessed in your class. If you are trying this for the first time, don’t feel that you have to modify an entire unit. You could begin by modifying tasks for one skill in a unit, and when comfortable, modifying for another.
Analyzing Task Complexity: a Model
Skehan’s framework (1998), mentioned in Chapter 4, provides a useful model for developing tasks/texts across more than one benchmark level. The framework helps analyze task complexity and supports developing tasks (and texts) to accommodate more than one benchmark level. Skehan outlines three factors that influence task difficulty: code complexity, cognitive complexity, and communicative stress (Skehan, 1998; Rossiter & Abbot, 2012). A brief description of each is given below.
CODE COMPLEXITY: the language required to complete the task.
Code complexity is interpreted at three levels:
Linguistic complexity and variety: Ideas are explained in simple or complex language.
For example, at a low level of complexity and variety, a task might have learners ask for a standard hotel room, and at a higher level, a room with specific requirements such as non-smoking or handicap access.
Vocabulary load and variety: Words in the task/text are simple or complex.
For example, at a lower vocabulary load with less variety, learners looking for employment information might read the Easy Reading Job Profile on the Alberta Government ALIS website, and at a higher level, the full Occupational Profile.
Redundancy and lexical density: Ideas in the text are elaborated (or not) using synonyms, paraphrasing, and examples. For example, at a lower level, learners interested in automotive topics might use a less dense (and lexically redundant) text like Cars for Dummies, while at a higher level, they might be able to access a more lexically dense computer manual on disc brakes.
COGNITIVE COMPLEXITY: the thinking required to complete the task
Cognitive complexity is affected by familiarity with the task and the level of processing required in the task.
Tasks are easier if learners are familiar with the following three features of the task:
Topic: Learners will likely find engaging in conversations about familiar topics like a typical work shift easier than engaging in conversations about unfamiliar topics like payroll procedures.
Genre: Learners will likely find reading or writing a personal note easier than reading or writing an unfamiliar genre like minutes from a meeting.
Task: Leaners will likely find listing work history on an application easier than the less familiar task, summarizing their workability skills.
Tasks are also easier if the amount of processing required to complete the task is eased, including these features:
Information organization: For example, learners will find listening to a sequential narrative easier than a narrative with flashbacks.
Amount of computation required: For example, learners will find giving directions from home to work easier than giving directions when there is a road closure.
Clarity and sufficiency of information: For example, learners will find reading a description of an apartment easier than a more abbreviated rental ad.
Information type: For example, learners will find sorting foods based the Canada Food Guide easier than making healthy food choices based on the Canada Food Guide recommendations.
COMMUNICATIVE STRESS: the conditions under which the task is completed
The task can be made more or less difficult by increasing each of the following conditions, most of which are self-explanatory, with appropriate conditions at individual CLB levels clearly described in the Some Features of Communication tables included in each Stage in the CLB (2012) document:
Time limits and time pressure
Speed of presentation
Length of text used
Opportunities for learners to control interaction
Type of response
Number of participants.
In reality, the adjustments in expectations between benchmark skill levels in the CLB framework are consistent with the features Skehan describes. We can see this in the changes in expectations between two tasks related to listening to the weather, one at CLB 4, and the other at CLB 5. At CLB 4 learners might be asked to listen to a very short, clear weather report to use the information to decide what to wear (CCLB, 2012, p. 11). At CLB 5, learners might be asked to listen to a weather report to inform members of a work crew about a delay due to weather (CCLB, 2012, p. 17).
The increase in expectations between the two examples is consistent with Skehan’s framework. An increase in code complexity is evident in the greater range of vocabulary needed in the CLB 5 task, an increase in cognitive complexity is seen in the greater computation required to use the information from the forecast to make a decision about adjusting the work schedule, and an increase in communicative stress is evident in the longer text the learners are asked to listen to.
Modifying Tasks to Accommodate Learners at Different Levels: A Process
Modifying your assessments to include more than one CLB level can be completed in a three-stage process.
Step One: the Unit
When adapting your assessment plan to accommodate diverse levels, you might start by identifying the additional level(s) you will be assessing in your class – using your class profile. You should then review your unit plan, ensuring that all components reflect the levels you are assessing for. The unit plan in Classroom 2 demonstrates how two levels can be accommodated in a unit plan.
Step Two: The Assessment Task
Tasks can be modified by using Skehan’s framework: adjusting the linguistic complexity, cognitive complexity, or communicative stress. As you review an assessment task and the material you are using, ask the question, “What adjustments make the most sense for this particular task?” You can modify your productive or receptive tasks based on your answer.
A productive task can be adjusted either up or down. For example, the CLB 2 productive task Make a Simple Food Order in a Restaurant would be set up to allow speakers to place the order face to face, supported by gestures and visuals, while using short phrases and some short sentences, with little evidence of connected discourse. The task might be modified up or down in the following ways:
Down to CLB 1: To adjust the code complexity, the instructor could require that learners use only a simple phrase or single word, such as “Coffee, please”; to adjust the communicative stress, the instructor could require that learners order only one item.
Up to CLB 3: To adjust the code complexity, the instructor might require learners to make requests in short sentences; to adjust the communicative stress, the instructor might increase the number of items that learners must order; to adjust cognitive complexity, the instructor might ask learners to include an appropriate opening and closing, or to respond to an unexpected question.
Modifying a receptive task starts with the reading or listening text. If you are assessing across two levels, you can often find a text that is appropriate for both levels. The Features of Communication tables for each stage provide helpful information to guide text selection. Also, the CLB glossary clarifies many terms (e.g., the difference between “non-demanding” contexts for texts at CLB 2, 3, and 4 and “moderately demanding” contexts at CLB 5-8). If you are assessing across Stages (e.g. CLB 4 and 5) or across three levels, you may need to adjust the text itself.
Once you have finalized the text, you can adjust the assessment task and assessment criteria as described above. For example, to modify a CLB 3 task, Read Simple Advertisements for a Product to Decide whether or not to Buy it, you might make these changes:
Down to CLB 2: To reduce communicative stress, the instructor might ask learners to read the first advertisement, to find details about price and store hours, but not to make product decisions (therefore also reducing cognitive complexity).
Up toCLB 4: To increase cognitive complexity, the instructor might include a second advertisement and increase the amount of computation required, asking learners to compare and choose between two products to make a decision.
Complete a class profile for your class. What patterns do you see in levels, both across skills and within skills? If you were to assess for two CLB levels in this class, which two would you choose? Would these two levels be the same for all four language skills?
Consider an assessment task you are currently using. If you were to adjust it to assess the second CLB level you’ve identified in question 1, what adjustments might you make? How are Skehan’s three features (code complexity, cognitive complexity, or communitive stress) reflected in these adjustments?
What advantages would you expect for the learners if you were able to assess for two CLB levels in your classroom?
Rossiter, M. & Abbot, M. (2012). Planning for and instructing multi level classes using the CLB. ATESL, Alberta Initiatives Gallery, November 13, 2012 [PPT presentation]. Retrieved from http://www.atesl.ca/node/245#.V4Wh2LgrLIU
Skehan, P. (1998). A cognitive approach to language learning. Oxford, UK: Oxford University Press.