Solano-Flores article

Solano-Flores: Who is given tests in what language by whom, when, and where? The need for probabilistic views of language in the testing of English language learners.

The article discussed the ways in which the current assessment methods are unsound and unfair for the English language learner (ELL). It mentions the many different ways in which the assumptions and methods are wrong when it comes to the ELL students. In explaining the sentence “who is given tests in what language by whom, when, and where?” The author explains how the assessments the ELL students have to take do not necessarily test what it is testing. The “who” portion of the sentence is the students themselves. The inconsistent definition of the ELL and their proficiencies makes their classifications inaccurate. In the “test” portion of the sentence, the author explained that the developing, adapting, and administering of the tests are difficult and that they are not implemented properly or consistently. In the “language” portion, it pointed out the fact that a language, like Yup’ik, has many dialects within its domain. Even in the English language there are dialects- village English, eastern dialect, so on. With each language there is a dialect. There is still a belief out there by some test developers, raters, etc. who do not understand, or know this. So the validity of the test is compromised because of the dialect variations. In the “whom” portion it is mentioned that there are times when the person is administering the test can sometimes effect the ability of the test taker wither negatively or positively. Even the developers of the test become questionable, because it is not clearly understood what they are faced with because they (developers) come from different experiences and knowledge. The “when” portion discusses the implication of when to test the ELL. The myths and misconceptions of language development and proficiency come into the picture. Just because the ELL is able to speak the second language doesn’t mean they are proficient in the second language. There is research that mentions that it can take anywhere from 3 to 7 years for the second language to be “proficient”. The “where” pertained to what works in one setting, doesn’t mean it will work in another because of dialect differences. This discussion led to the probabilistic views of language in the process of ELL testing (p. 193). In this view, there is no set determiner of what is being tested. It “approximates” the item or items to be tested. In this view the theory of generalizability (G) theory was explained.
In reading this article it made more and more sense to use authentic testing for the language learner. In authentic testing, the main purpose for me would be the student. What they are learning, where they are having difficulty, and go from there. With all the negativity with standardized state and national tests have for the language learner, it makes sense to use the authentic assessments.