You make a good point, but I wonder if part of the answer might be to expand the definition of "assessment." Currently, NCLB employs traditional multiple-choice standardized assessment. I'm not sure if these types of assessments are truly measuring the kind of learning we most want for our children. As it stands, children learn a large number of discrete facts because this is what is required for the assessments. What about connected learning and meaning? Perhaps, we should try performance-based assessments, portfolios, video, etc. There is, indeed, increased subjectivity in these kinds of assessments, but meaningful learning is quite subjective overall!

Good points and I could not agree more. Actually, several states have moved to performance standards beyond traditional skills.  I think one concern about some of the porfolio based assessment directions is the extensive time required by teachers to conduct. Also, authentic and porfolio assessments introduce more subjectivity in judgement. Of course, this can be good and bad.

I agree that assessment is essential for identifying areas in which students need help and also for encouraging accountability.  However, I think that two perspective shifts are needed to make assessments effective and fair.  First, I think assessments should be presented to students as measures of learning, not as measures of their individual worth  -- high stakes, high pressure environments in which failure entails a huge loss of self-esteem practically set students up for poor performance.  Second, I think that when looking at the results of assessments, it's important to look for where the school, teachers, and community can improve, rather than merely pointing out how students have not met expectations.  Taking these perspective shifts to heart means using assessments as tools for change rather than tools for blame.

Lindsay, there are proposals from many of the school systems that captures part  of the spirit of your two points. Actually, Time magazine last week published an excellent article on No Child Left Behind and a discussion of this very point. Currently, NCLB requires  reporting a school's progress on the snapshot of the students' performance on the state's high stake annual assessment. This high pressure test is what you and Laura oppose. I agree. I do like the idea of tracking the annual progress of each student not the global statistics of the group. A school is penalized if students are not at grade level. However, the more important measurement should be based on the students growth. How much growth did students achieve. We can expand the assessment to include Laura's expanded notion of the "what" we measure. Example, students in a school may gain 2 years growth; however, the school is still classified as not "working" if the school as a whole is below grade level. Measuring individual growth of students, classes and schools is the key for assessing the degree of success.  The debate over global statistics vs. the student "growth" model is the current hot point of the debate.

I agree with your ideas on expanding the types of assessments and the things we assess, but it would be very interesting to drill down to issues such as scale, minimizing subjectivity in assessment, cost of assessment, national standards vs. state standards, and the issue of states achieving success on NCLB because they have severely water down the standards. As an example, the  4th grade literacy Mississippi standards in literacy  is equivalent to the 2nd grade literacy standards in CA and in Mass.

Assessments should measure the growth of each individual child from year to year.  Especially those students with IEPs.  Just because that can't pass the "test" doesn't mean there wasn't any learning going on.  The tough question is how to do that.