The measurement of one's linguistic competence (I prefer to used "linguistic proficiency" rather than "competence" to avoid the old Chomskyean terminology of Competence / Performance which is intended for an idealized speaker and not a real one) is not an easy task. Given the influence of context and intentions on the actual manifested proficiency of a speaker, one test or set of observations doesn't effectively provide sufficient data to make such claims. It was Jim Cummins (among others) that made this point when he was trying to discuss why one could not use a specific test instrument or a focus on a particular task or context to make a blanket statement about one's proficiency. This is where his BICS vs. CALP distinction was first employed....unfortunately, too many people picked up on this dichotomous distinction and NOT on the main point...which was... if you want to see how effective one is in a particular setting or context, you HAVE to look at their performance in that context or setting. There are simply too many variables that impact proficiency (e.g., context, intention, experience, affect) to be able to make a blanket statement.
That said, I look at one's proficiency in a real task (social, academic, epistemological) by looking at actual performance in that task and asking the questions of interest to me....if I want to look at overall meaning-making in the classroom, I look at how language is used in the task or the classroom....for example, comprehending lectures, taking notes, entering into discussions, asking for clarification...I let the context and the needs of the setting dictate what one is looking at. If I want to look at grammatical proficiency in conversation, then I look at the effectiveness of the grammatical components the person uses during conversation and ask whether these were sufficient to accomplish the tasks at hand...communicating for a particular purpose. In these situations, I use authentic data via observation or via transcribed oral text. I wish it were easier....but...is it is not. See the attached document for more info.
This happen by either recording samples of speakers communication or by observation. If their speech grammatically correct, this means it is linguistically competent. And if their speech dilever the meaning Well, it is communicatively competent.
The measurement of one's linguistic competence (I prefer to used "linguistic proficiency" rather than "competence" to avoid the old Chomskyean terminology of Competence / Performance which is intended for an idealized speaker and not a real one) is not an easy task. Given the influence of context and intentions on the actual manifested proficiency of a speaker, one test or set of observations doesn't effectively provide sufficient data to make such claims. It was Jim Cummins (among others) that made this point when he was trying to discuss why one could not use a specific test instrument or a focus on a particular task or context to make a blanket statement about one's proficiency. This is where his BICS vs. CALP distinction was first employed....unfortunately, too many people picked up on this dichotomous distinction and NOT on the main point...which was... if you want to see how effective one is in a particular setting or context, you HAVE to look at their performance in that context or setting. There are simply too many variables that impact proficiency (e.g., context, intention, experience, affect) to be able to make a blanket statement.
That said, I look at one's proficiency in a real task (social, academic, epistemological) by looking at actual performance in that task and asking the questions of interest to me....if I want to look at overall meaning-making in the classroom, I look at how language is used in the task or the classroom....for example, comprehending lectures, taking notes, entering into discussions, asking for clarification...I let the context and the needs of the setting dictate what one is looking at. If I want to look at grammatical proficiency in conversation, then I look at the effectiveness of the grammatical components the person uses during conversation and ask whether these were sufficient to accomplish the tasks at hand...communicating for a particular purpose. In these situations, I use authentic data via observation or via transcribed oral text. I wish it were easier....but...is it is not. See the attached document for more info.
Collecting data on linguistic competence involves several considerations. First, you have to identify the learners' characteristics; namely, age and background knowledge.Second, you should decide on the type of approach to identifying linguistic competence by adopting either a norm-referenced or a criterion referenced testing method. Third, you can increase the validity of measurement by integrating both testing methods. In point of fact, you can initially administer a proficiency test and then engage the targeted participants in performing a mastery/non-mastery task such as an interview or composition writing.