Discussion about this post

User's avatar
William Lynch's avatar

William T. Lynch, PhD. March20, 2024 ... Yes, the Federal Department of Education should not be eliminated, and it is equally true that the State and City Governments should be held responsible for providing a proper education to all citizens. After all, about half our State income and property taxes are already assigned to Education, and more money has never been the solution. For thirty plus years the metrics employed for evaluating student and cohort progress have been totally wrong. Point scores, and their algebraic differences, do not measure progress. But relative knowledge, and absolute relative gains, can be quantitatively measured, with comparisons valid over decades. The national plan for NCLB was noble, but was destroyed by States, politicians, and teachers’ unions who guaranteed a false success by means of horrible metrics. Individual states also wished to avoid their deserved embarrassment. Only the “Nation’s Report Card” (NAEP) is left as a common standard. The latest suggestion for a replacement for NCLB that allows the States to conduct their own tests will also offer a common linkage via the common nationwide NAEP tests was rejected by a Congressional Committee.

The recommended set of normalized metrics that have been proposed will give the same results for all “good” tests that profess to cover the same topic and grade. But it is required that all students being compared have taken at least one common test. (NAEP, e.g., meets that requirement.) The range of “difficulty” of the tests does not matter. For any pairing of cohorts (ethnic, poverty, Lunch vs. NFL) these performance and gap metrics have been consistently reliable: District tests, end-of-year State tests, NAEP, and even SATs (for its common sets of students).

They also provide more obvious (and realistic) goal setting for year-to-year progress (even if the tests are being changed every few years). For example one such metric is the answer to the question: “What percent of students in lagging cohort B have scores below the 25-percentile score of the control cohort?” One would like that to be 25%, but it more likely to be 60% to 65%. And it has not yet been proven in the classroom, but it could be (should be) demonstrated (as it is on paper) that, when students are differentiated in their classroom assignments as belonging to the lowest 20 percentile, the highest 20 percentile and an intermediate 60 percentile – with a significant overlap that is the annual responsibility of the Principal to adjust – the burden on the teacher is greatly reduced AND the progress of each category will be expanded enormously. A full disbursement of 100% presents a teacher with students having a current confirmed knowledge (that does not mean aptitude) range of about 30. The three-way split (with overlaps) will give each teacher a more manageable range of about 5. It may be desirable to wait until the fifth grade to initiate the splits. Common accord between Principals and teachers is the trigger that will make this work, not political intervention. For us to show progress in the world we must introduce these changes.

Expand full comment
Clint Laird's avatar

once again John Droz applies discipline (i.e. critical thinking) to his analysis of complex issues - particularly the disappointment in our public education system.

Regarding Mr. May's 2 reasons below:

1) empowerment of teachers unions was not addressed. If that's a reality, it is one that can be eliminated via reform.

2) see John's #6 on "correlation does not prove causation"

John's postings on education (and Critical Thinking in particular) are well worth the time and intellectual intention to read .... thoughtfully. John adds to my understandings every time I invest the time to read & consider his posts. Thank you John

Expand full comment
57 more comments...

No posts