viernes, 21 de mayo de 2010

1st Piece of Work - Is Communicative Language Testing an Attainable Goal?

Some guidelines to work on the article by Miyata-Boddy and Langhan. Choose two of the questions and send your comments:
  1. Why do you think we are reading this article now, after reading McNamara's introduction to testing, discussing Oller's views and analysing the activities in (young) Jeremy Harmer's text book?
  2. McNamara starts chapter 2 highlighting the connection between the types of tests and the different test designs with the view of language (and language learning) the test developer/teacher holds. Does this article support McNamara's assertion? Explain.
  3. Why do you think the authors devote an important section of their article to the description of communicative competence models? Work still needs to be done in the field of communicative language testing - can you identify the areas that still need to be illuminated.
  4. Can you list the characteristics of communicative language testing, according to this article? It is clear that the authors are more concerned with tests that are massively administered. Do you think these characteristics they postulate have implications for classroom assessment?

10 comentarios:

  1. Hi Flavia, I`ve tried to answer ners 3 and 4.. I don`t know if they are ok...I had problems in interpreting the last question: "do you think these characteristics have implications for classroom assessment?" well..here it goes!

    3) They devote an important section of the article to the description of communicative language competence models because, according to Spolsky (1989:140), when we test a student we are measuring that person`s knowledge of and proficiency in the use of the language. Thus, testers need to know what is knowledge of and proficiency in the use of a language, they need to know, what does it mean. If testers do not have that clear in mind, they will not know what to test and how to test that. I mean, they would not be able to stablish the "criteria" for testing.

    4) Features of communicative language test:
    - sholud have high "content validity", that has to do with "authenticity" of task.
    - need to be "context-specific"
    - should be assessed "qualitatively" not quantitatively.
    As regards the implications of these characteristics for classroom assessment, I think that, although this kind of testing seems to be really demanding on the part of the teacher (design of tasks, give a context, etc...)
    it is worth implementing it because sts are asked to "use" the language for communication since they put all their knowledge into practice and use the necessary strategies to convey meaning.

    that`s all for now...
    CU,thanks!!
    virginia.O

    ResponderEliminar
  2. Hi Flavia! I´ve chosen to answer the questions number 2 and 4. I don´t know if what I wrote is OK...

    2) According to what I have read, Nick Miyata-Boddy and Clive Langhan support Mc Namara´s assertion because they show us how testing has progressed. "It has passed through eras when reliability and objective testing were dominant to the period today when testers are more interested in how a candidate is able to use his/her knowledge of language in a communicative situation than a demostration of the knowledge in isolation". (Miyata-Boddy and Clive Langham 2000:81) In my opinion, the former (reliability and objective) refers to what Mc Namara describes as "Discrete point tests" (which focuses on candidates´knowledge of the grammatical system of vocabulary and of aspects of pronunciation and tends to atomize and descontextualize the knowledge to be tested and to test aspects of knowledge in isolation) while the latter has to do with "Integrative and pragmatic tests" (characterized by integrated knowledge of relevant systematic features of language with an understanding of context).

    4)The characteristics of communicative language testing are:
    - "high content validity": it has to do with "authenticity".
    - "context-specific": I have a doubt here. I understand that it has to do with an attempt to reflect the real life situation taking into account the use of genuine texts. Is it right?
    - they should be assessed qualitatively and not quantitatively. They should reveal the quality of the testee´s language performance.
    As regards the second question, I think that these characteristics have implications for classroom assessment. In other words, I believe that everything is connected and that the more clearly and specifically you understand how students are learning, (gathering information about what they know, are able to do, etc) the more effectively you can teach and evaluate them .
    As Virginia O., I´m not quite sure about this last answer because I had difficulties in interpreting the question.

    That´s all.
    CU, thanks!
    Virginia L.

    ResponderEliminar
  3. Hi,
    So far I´ve chosen question number 2.
    In my opinion, this article fully supports Mcnamara´s assertion since both accurately reflect how testing has gradually changed. Even though tests can serve pedagogical purposes, there are qualities of tests that both texts share: reliability and validity. As you may have noticed, we have similar ideas.
    Another important idea to be added is the concept of "task" that can be defined as "any activity in which a person engages, given an appropriate setting in order to achieve a specifiable class of objectives" (Carrol 1992:8). Tasks take into consideration the cognitive process of the learner as well as the context related to real life situations. They started to be used in the 70´s in which knowing a language is more than knowing its rules of grammar. This is the opposite of the behaviourist view (which would be the case of discrete point tests) which puts emphasis on habit formation through extremely guided activities and in which the product is only considered.
    I strongly believe that the concept of task is essential and that is the reason why I have made a brief comment on this.

    Tomorrow I´ll try to post number 1 because noone have worked on that one.

    I hope this is Ok.
    See you.
    Gise

    ResponderEliminar
  4. HI Flavia, I would take the shot and answer number one, hahaha!!!!
    I think that we are reading now this paper because so far we were introduced to different kinds of tests and how the changed over time. We saw how the discrete-point test and others have their limitation, and that nowadays the best option (because include other aspects of the learner, not only the competence but also the performance of the learner) is the communicative test. However, Miyata-Boddy shows us that even communicative tests have flaws and they need revision. In this paper, Miyata-Boddy states in full account, the problems that the communicative tests have and also how these problems have been addressed. For example, the subjectivity at the time of assessing and which rating scale is best to use can be solve now by “given sufficient training and standardisation of examiners to the procedures and scales employed” (Miyata-Boddy, 2000).
    I don’t know if this answer is Ok, but I tried.

    ResponderEliminar
  5. Hi girls! Thanks Virginia, Virginia, Gisela and Leila for your contributions! They are very valuable and I hope they promote more and more participation on the part of your classmates.
    A few comments:
    Leila is quite right about the reasons for reading this article now, that we know what discrete point and pragmatic tests are. She highlights the fact that communicative tests have their shortcomings as well, and that work need to be done to overcome them (related to question 3, in fact). Subjectivity is one of the controversial issues about communicative testing that she mentions, and it is worth giving this concept some thinking time.
    There is a strong connection between subjectivity and the two concepts Gisela highlights: validity and reliability. It’s true that there is a clash between these three concepts in the discrete point paradigm, but we need to acknowledge that reliability and validity have been redefined as new teaching and testing methodologies appeared.
    On the same line of thought, I totally agree with Virginia L. when she says that reliability and discrete point testing are closely related. However, I cannot see the link between objectivity and pragmatic testing. Can you enlarge on this, Virginia?
    By the way, both “Virginias” are quite right about the characteristics of communicative language testing. I’ll give other students the chance to answer this question, and then I’ll give you my opinion on the implications of communicative testing for the L2 classroom.
    I do agree with Virginia O. when she says that if it is communication what we want to assess, we should then be aware of the many competencies involved when people use the language (discourse, socio-cultural, strategic, linguistic, …). This awareness should illuminate our teaching, the contents we select and the assessment methodology we adhere to. This explains why the authors devote an important part of the article to the models of communicative competence.
    This is it, girls. Your turn again!

    ResponderEliminar
  6. Hi! Regarding the second question and taking into account what I quoted from the paper in the previous comment (Miyata-Boddy and Clive Langham, 2000:81), what I tried to say (maybe I didn´t express correctly) is that “the former” refers to the first era when reliability and objective testing were dominant, that is why reliability and objectivity are closely related to discrete point tests, while “the latter” refers to the period today when testers are more interested in how a candidate is able to use his/her knowledge of language in a communicative situation than a demostration of the knowledge in isolation. This is, in my opinión, what is related to integrative and pragmatic tests, not objectivity.
    I hope you understand what I´m trying to say.
    Thanks!
    See you.
    Virginia L.

    ResponderEliminar
  7. Hi Flavia, I will try to answer #4.

    “Since communicative language testing inteds to provide the tester with information about the testee´s ability to perform in the target language in certain context-specific tasks, tests should:

    - have CONTENT VALIDITY (related to AUTHENTICITY OF TASKS and RELIABILITY) and CONTEXT-SPECIFIC, i.e the sample of language collected and the tasks the candidates is called upon to perform should be as representative as possible of the language and skills needed to function in the real life context.”

    - “ Have an element of unpredictability, i.e the processing of unpredictable data in real time is a vital aspect of using language.” I´m not quite sure whether this is a characteristc or not.

    - “Be assessed QUALITATIVELY rather than QUANTITATIVELY”

    However, these characteristics are not always easy to fullfill since “only a small sample of the testee´s language can be collected and that however realistic the task may be intended to be, the testee´s performance will inevitably reflect the fact that s/he was performing under test conditions.”
    As regards the second question I think that as Virginia O mention, this kind of tests are worth implementig not only because you get ss involved in real tasks but also because they promote ss self-steem, motivation and positive attitudes towards learnig and the target language.

    ResponderEliminar
  8. Hi Flavia !!! So far, I will tray to answer question number 4:
    According to this article, there are three characteristics of communicative language testing:
    · High content validity: the test has to reflect the situations in which the individual is involved in real life.
    · Context-specific: the language and skills used by the individual in real life will need to be reflected on the sample of language collected and the tasks the individual is called upon to perform.
    · Quality over quantity: tests should reveal the quality of the testee’s language performance.
    As regard as question number 2, I consider that these characteristics have implications for classroom assessment. As Virginia O. said, even tough these tests demands several things from the teacher, it enriches the students’ learning processes of the language as they make use of all their knowledge to express themselves in a meaningful way in situations that are similar to real life.

    ResponderEliminar
  9. We (Marcela and Vanesa) have been working together and we'd like to add some comments to questions number 1 and 3.
    As regards question number 1, we agree with Leila's opinion. We go on to say that we are reading this article after reading McNamara and Oller's in order for us to notice the contrast between different kind of testings (descrete point tests-integrative tests vs communicative competence tests).
    Considering questions number 3, we think that the author devotes an important section of the article to the description of communicative competence models because, on the one hand, as Virginia O and Flavia say, the author wants to make us aware of all the competencies we should take into real consideration when we test communication.
    And on the other hand, we think that the author also wants to set this kind of test (communicative test) apart from other forms of testing in which testers assess quantitatively rather than qualitatively.
    That's all folks!! (so far...)
    Vanesa-Marcela

    ResponderEliminar
  10. Hi there!
    Here I am... better late than never! Haha!
    I guess there’s not so much to be said, but anyway, here I go. (Questions: 1 & 2)

    1- I think we’re reading these connected texts to realise that language view and test development go hand in hand.
    As McNamara (2000) presents us, language can be seen:
    • As a set of behaviours to be learnt,
    • As a system (sts have to be aware of the rules to be able to apply them),
    • As communication, i.e. we have to pay attention to use AND usage.
    If we are aware of this key relationship, we’ll be able to prepare valid tests (including what has to be included according to our view of language and the way in which we have taught it). I don’t mean that we’ll be expert test developers, but this theory will help us take into account all kinds of tests at the moment we need to prepare one or combine them.
    It’s worth highlighting that not only do McNamara and Miyata-boddy and Langhman show us the characteristics of different tests, but they also emphasize how useful communicative language tests are when we want to evaluate sts’ competence and performance. We, as teachers, have to be eclectic and combine all the methods and approaches needed to test our sts.
    Now that I’m writing about the relation between language views and the types of tests, (Ner 2) I state that the article by Miyata-boddy and Langhman does support McNamara’s assertion. As some of the girls have written, both texts describe the processes that testing has gone through to come to the following conclusion: “It has passed through eras when reliability and objective testing were dominant to the period today when testers are more interested in how a candidate is able to use his/her knowledge of language in a communicative situation than a demostration of the knowledge in isolation" (Miyata-Boddy and Clive Langham 2000).

    Well, that’s all ☺.
    Seeya!
    Naty

    ResponderEliminar