Data is used as evidence of learning. It is all a matter of judgment and how the test is used and if used for the test's purpose. Test validation is when investigation undergoes to see if the procedures were achieved by the judgment concluded, so much like a trial and presenting information to a jury in court. The book gave an example of OJ Simpson's trial. I think the format that you use to see if a test's criteria is met is as important as putting or keeping an individual from jail time.
The validity of assessment procedures and what the test measures should be revisited and revised to meet the needs of your students in any testing situation, whether it be in your classroom or a high stakes test like the SBA or the HSGQE test that the State of Alaska uses to judge schools to see if they are making adequately yearly progress.
The test developer has a huge role in determining the passing and qualifying of the test before giving it to others to use to measure achievement or ability levels.
Test validation should go through an analysis to see if the data it judges is valid and is used for the purposes of the test. Of course, the level that the test is validated will all be based on funding and whether there is money for the test developer to revise the test to make it as valid as possible. This of course takes time and lots of data gathering and willingness to follow through on testing materials.
I see the point of making sure that students are taking valid tests in language. It makes me think of the purposes that tests are used for and how some might be misused to "validate" data.
Back to Work
14 years ago
2 comments:
Hey Guru Pitka! I agree with you that tests need to be tested before passing them onto the school and before the language learners' use. I also agree that the process to test these tests is a long process and should be revisited often. You make excellent points here and ones that I am hoping test makers keep into consideration. It makes me look back at my tests and little quizzes that I use and thinking I should go back to see it they are valid. Maybe I would need to add to the testing process and not use only paper and pencil formats all the time. I beginning to find that recording data is another way to test. I never really recorded as a means of testing, but that can be a mode, as I am learning it's great to me multimodal and the idea of multiplism is a fresh concept. Hey!
Mae,
The simple answer to classroom validity is ...Does the assessment measure what you want it to measure? Does the assessment match the instruction in your classroom?
The challenge is helping the community members and district and state to VALUE these ongoing assessment measures as much as they do the standardized ones.
Marilee
Post a Comment