Language Learning and Assessment
This essay looks at how a variety of testing items can align to the desired results put forth in a written curriculum. The educational context of an English as a foreign language classroom fosters the creation of understandings, knowledge, and skill sets as an inclusive set of curricular aims. Norm and criteria-referenced tests are discussed in terms of how each can benefit from assessing English language learners’ communicative competency as well as their academic English skills. It was determined that both types of tests are equally important in assessing English language learners and that the best approach to implementing such tests is through a community of practice that promotes shared and reflective teaching practices in a risk-free educational environment.
Language learning and assessment: setting behavioral objectives through the development of understandings
Assessment in language learning often focuses on behavioral objectives that are based on skills (i.e., grammar usage, phonetic distinction, lexical ability, etc.). How English language learners communicate is often measured in terms of fluency, accuracy, and perhaps sociocultural elements to language as well. Limiting assessment on skill-based behavior runs the risk of overshadowing the potential for language learners to achieve higher levels of achievement as Bloom states in his taxonomy as follows: “analysis”, “synthesis”, and “evaluation” (as cited in Kubiszyn and Borich, 2007, p. 95). Wiggins and Mctighe’s put forth a slightly different notion of establishing learning outcomes through their pursuit of “six facets of understandings: a) explain, b) interpret, c) apply, d) perspective, e) empathy, and f) self-knowledge” (Wiggins and McTighe, 2005, pp. 85-102). The six facets of understanding contrast Bloom’s taxonomy in that the former is not hierarchical and are not limited to only the cognitive domain. Indeed, the affective and psychomotor domain are addressed as well through an emergent, phenomenological perspective (i.e., teachers facilitate learners through various performance verbs that are not specific to only one or two facets of understanding as opposed to being determined prior to instruction). In determining the evidence required to infer what English language learners should understanding, know, and be able to do, a combination of “selected and constructed response” (Popham, 2008, p. 115) test items are needed.
Before planning assessment test items, the desired results, or classroom objectives, must be determined. The desired results can be expressed in terms of understandings, knowledge, and skill sets. In other words, assessing the English language learner (ELL) builds not only on a certain skill set (i.e., pronunciation and grammar usage), but also some cultural knowledge and understanding of sociocultural practice. Taking a typical topic from a level I English course as an example, an understanding might be as follows: The English language learner will understand that the manner and way in which people greet and introduce each other depends greatly on the social context. In order to achieve this understanding, ELLs will need to know under which social contexts speakers use formal and informal register, and they will need to be able to use the present tense form of a variety of verbs and appropriate vocabulary in order to successfully introduce themselves and others as well as greet both friends and strangers. An example of an instructional objective that is based on what an ELL should know and be able to do is as follows: After reflecting on a give social context, the ELL will be able to effectively greet and introduce someone using the appropriate use of language in a way that is understood by a native or near native-like speaker. When assessing these desired results (i.e., understandings, knowledge, and skill set), several types of assessment measures are necessary to assure that the evidence the ELL provides is valid, reliable, and non-bias.
Assessing desired results include both norm and criterion-referenced tests. Norm-referenced tests (NRTs) include in-part multiple-choice, true-false, and matching test items. Continuing with our example, a multiple-choice question that attempts to assess our instructional objective might be the following: When greeting a good friend at a party – one you typically see on a daily basis at school – all of the following are acceptable introductions except:
a) Hey, man. What´s up?
b) What’s going on?
c) How´s it going?
d) Excuse me, how are you tonight?
A true-and-false question measuring the same desired result:
When greeting someone, the utterance, “What´s up, man?” is not used to address a female. (T/F).
An example of a matching exercise follows: Match the possible opening greetings (1-5) with the most appropriate follow-up response (A-E). Write the appropriate response in the space provided.
1. Hey, what´s up man? _____ A. I´m alright, dude. But I just had a fight with my girlfriend.
2. Hello Sir (a stranger), how are you? _____ B. Yeah, it sure is.
3. My name is Bob. What’s yours? _____ C. Not much dog. What´s new?
4. How are you today, Bob (your friend)? _____ D. Tim.
5. Nice weather we´re having today? _____ E. I fine thanks, and you?
Answers: 1. C; 2. E; 3. D; 4 A; 5. B
With NRTs, answers are objective; that is, there is usually only one right answer. Questions are typically reduced to assessing discrete facts and knowledge usually associated with the lower three cognitive objectives: knowledge, comprehension, and application (Kubiszyn and Borich, 2007).
Another type of assessing desired results is through constructed-response tests (CRTs), specifically essay writing. Kubiszyn and Borich (2007) make a distinction between two types of essay responses: extended and restricted. An extended essay “can vary from lengthy, open-ended end-of-semester term papers or take-home tests that have flexible page limits (e.g., 10-12 pages or no more than 20 pages)” while a restricted response essay is “restricted to one page or less” (p. 136). An extended essay typically is most appropriate for assessing understandings whereas restricted essays are more appropriate for knowledge or non-contextual, discrete facts. To assess the greeting and introduction example, the following essay could be used to assess an understanding: You were contracted by a tour agency to provide a written guide for foreigners to the United States in order to help them get around a new city. Since the foreigners speak little-to-no English but do read a good amount of English, you must include in your written guide several different social contexts and dialogs in order for the foreigners to have an idea as to which language is most appropriate. They will need to know the different registers (formal or informal language use), non-verbal communication, key vocabulary terms, and other cultural norms for each given social context. Each dialog should have three-to-five turns in order to properly describe the language that is to be used, and standard English should be used when completing the guide, playing close attention to grammar, spelling, and punctuation. The following essay could be used to assess knowledge or non-contextual facts: Imagine you are at a party with friends, but there are a few people who you don´t know. Write out a 10-turn dialog that takes you through a conversation as you introduce yourself, both to friends and to strangers. Formal and informal language as well as side-notes that specify non-verbal communication should be included in your dialog. Your writing should include correct grammar, spelling, and punctuation. Your entire dialog should not exceed one page. Contrasting NRTs, CRTs allow for a variety of possible answers and provide the ELL a level of choice in the learning process. Although CRTs tend to be more subjective, using “good” rubrics can help make assessing CRTs as objective as possible. A “good” rubric necessitates groups of teachers working together in determining what criteria are most appropriate for a given collection of test items.
Assessment in language learning extends beyond the typical skill-based emphasis seen in the past. Instead of only testing language use (i.e., communicative competency), testing for understanding and knowledge can also be incorporated into the learning process as English language learners acquire higher levels of academic English as well. Norm and criteria-referenced tests provide both objective and subjective testing items respectively, thus achieving higher levels of cognitive, affective, and psychomotor development. In order to effectively achieve this higher level of development, assessment must be aligned with the desired results, or curricular aims in a way that promotes a community of practice that promotes a shared and reflective teaching practice in a risk-free educational environment.
Kubiszyn, T. and Borich, G. (2007). Educational testing and measurement: Classroom application and practice. Hoboken, NJ: Wiley and Jossey-Bass Education.
Popham, W. (2008). Classroom assessment: What teachers need to know. New York: Pearson.
Wiggins, G. and McTighe, J. (2005). Understanding by design. Alexandria, VA: ASCD. ´