Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 
You Are Here: Read an Article > View All Posts for the Article > Read a Post
 
Read a Post for Tenuous Findings, Tenuous Policies
 
Reply to this Post
 

Testing and Interpretation

Posted By: Fred Welfare on November 1, 2014
 
Tests are indicators which are more useful for program evaluation than for comparison. But, comparisons are fun. Who's #1? Test scores, like grades, are ambiguous. What do they really mean? There should be norms: what is above average, average, and below average, at least. If a school district or nation produces scores that are skewed or contain outlyers, then qualitative research should address that statistic. Why a nation is producing very high test scores may not be the same reason why another nation is producing similarly high scores. An investigation has to be made as to what the scores mean and how they are achieved in each particular region. This should not be understood to claim that testing is irrelevant or so besmirched with problems of validity as to reject their significance. Tests have different purposes; some tests establish program effectiveness, other tests select the best candidates for a particular task. Some tests are meant to indicate minimal qualifications and other tests separate out the most promising. Testing, and its antecedent 'studying,' should not be taken for granted or rejected as has become common in the US. Testing is meant to indicate competence and should be taken seriously. Where high school dropout rates are high, test scores are low which should propel program evaluation and policy change. Students however have to learn to test, how to study for tests and how to take tests, and how to get their intellect in shape to address testing. Formal test averages, or pass-fail markers, in a single school's department, say math or english or social studies, speaks volumes about the teachers' abilities and the students' backgrounds. I see no reason why we can't say the same thing for entire states and nations. Glaring statistical lacunae can be addressed in the next round. But, there is obviously a difference between the top and the bottom of a range of test scores: the bottom scorers need to address their problem of effectiveness. However, the most contentious issue is the cut-off score: where is the line between minimal competence and incompetence? Do our tests test for this difference? There should be no disputing what a failing score means in any country.
Thread Hierarchy
 Testing and Interpretation by Fred Welfare on November 1, 2014
     
    Member Center
    In Print
    This Month's Issue

    Submit
    EMAIL

    Twitter

    RSS