By Robert Oliphant Columnist EdNews.org

"Toward a Growth-Centered Assessment Model," a recent white paper from Pearson Educational Measurement (April 2006, and reprinted in EdNews, 12/26/06) can fairly be called a Declaration of Educational Independence for American students and their families.Written by Thomas H. Fisher and Jon S. Twing, it asserts as self-evident truths that today "statewide assessment and accountability are seldom stable over time," along with a need for a "standard measure of performance" capable of "withstanding the changes" that are bound to come crashing in "as governors, legislators, and education officials leave office."

Fisher and Twing, paralleling Jefferson's catalog of abuses by George III, catalog the many conflicting interpretations of our currently popular Value-Added model for educational measurement.And just as Jefferson in effect opened the door to the framing of our Constitution, they open the door to nominations for new performance standards by asking that they be transparent, verifiable, and "not overly complex" the three classic requirements for a "model," as stated in the International Encyclopedia of Unified Science, edited by Otto Neurath and Rudolf Carnap.

Transparency. . . . As far as transparency goes, the American dictionary is literally and figuratively an open book to all of us from age seven to age seventy and beyond.Like Tolstoy's happy families, our four major college dictionaries, now available via CD ROM, are fundamentally the same in presenting roughly the same 70,000 lower-case headwords and 200,000 definitions.

For test designers, each of these 200,000 headword-definition combinations has always been fair game, especially for those who devise crossword puzzles and spelling bees.So the American college dictionary as an established and authoritative institution is clearly a prime candidate as the locale for a standard measure of performance strong enough to withstand year by year politically and professionally motivated assaults upon it, including those of wild-eyed spelling reformers from Theodore Roosevelt on down to the present.

Verifiability. . . . Spelling bees have always relied on the American dictionary as their performance standard in verifying answers to definition-first questions, e.g., Please spell the word that is pronounced as /par"euh laks'/and defined as "the apparent displacement of an observed object due to a change in the position of the observer."The same has also been true of definition-first crossword questions, e.g., (a) Please spell the 4-letter word whose definition is "a house, apartment, or other shelter that is the usual residence of a person, family, or household," or (b) Please spell the 6-letter word whose definition is "an injection of a drug or medicine; shot."

Most of us, not just experienced crossword puzzlers, will agree that question (b) is substantially more difficult than question (a).If we track (b) down in the CD ROM version of any of our major four college dictionaries via its number of letters (represented as ??????) and its definition, we will find the above definition listed as number 4 under the headword NEEDLE, which is far less familiar to us than the definition listed for (a), which appears as number 1 under the headword HOME.

Practically considered, then, our college dictionary can also give any student or parent verifiable and quantifiable information regarding the potential difficulty of any definition-first question out of the 200,000 possible questions a spelling bee host or crossword designer might choose to design.Even more important, this kind of quantifiable information can be translated into a practical measurement scale via which any student or parent can measure difficulty levels and match them up with instructional levels, ranging from first-grade questions up to those suitable for professional school, e.g., medical and botanical single-definition terms.

By way of illustration:Education News 12/19/2006 presents a 5-element scale-formula for representing this kind of dictionary information in the article, "CD ROM Dictionaries, High Stakes Testing, and the Need for a Vocabulary Question Familiarity-Difficulty Measurement Scale."Simplified to three elements, the formula can be here stated as "Question difficulty depends upon the target word's number of letters plus the ordinal position of its definition (1st, 2nd, 3rd, etc.

, going on up to over 100), which can then as a whole be divided by the total number of definitions listed for that target word." . . . or, more succinctly, Difficulty equals Letters plus Ordinal Position divided as a whole by Definitions (Diff = (L+O)/D/

If we use the Random House College Dictionary and apply this formula to the target word HOME, it takes shape as (4+1)/24 and produces a difficulty rating of only .21 as opposed to NEEDLE, where it takes shape as (6+4)/17 and produces a higher difficulty rating of .59.

If we apply our formula to the single-definition medical term MYOCARDIUM "the muscular substance of the heart," it takes shape as (10+1/1 and produces a much higher rating of 10.Since this formula and its results are based upon verifiable dictionary evidence, it deserves careful consideration right now as a growth-centric assessment tool in school vocabulary programs and even anti-Alzheimer's program in senior centers that emphasize crossword puzzle competitions in their therapeutic programs.

Non-Complexity. . . . Call it "elegant" or "parsimonious,"the most practical adjectival equivalent of non-complexity is just plain "cheap."This is to say that for under $20 any student or parent can buy the CD ROM version of a national-standard college dictionary (American Heritage, Merriam Webster, Random House, or Webster's New World), create definition-first practice questions at the rate of two a minute, determine each question's level of difficulty, and design a practical program for raising his or her level of definition-first vocabulary achievement — all these at a per unit cost far below the expensive statistically normed tests that politicians and professional educators now argue over.

This assertion should not be taken to mean that educational measurement as a discipline should be replaced by do-it-yourself program monitoring.Far from it.What it means is that we should allocate our educational measurement resources in a nationally more productive manner, saving money on vocabulary testing that can then be used for more reader-monitored writing samples and multiple choice tests in mathematics and the sciences.

And while we're at it, why not replace "psychometrics" (a spooky echoes of Alfred Hitchcock) with terms like "mathemetrics" for the quantitative sciences and "leximetrics" for disciplines for whom "the limitations of language are the limitations of my world," as Wittgenstein famously put it.

TO CONCLUDE. . . . If not Pearson Educational Measurement itself, Thomas H. Fisher and Jon S. Twing have in my view certainly produced a revolutionary call to action by asserting America's need for a "standard measure of performance" that will — just like the National Bureau of Standards — protect us from the hot breath of political ambition and the cold trickeries of commercial greed.

My own response to that call naturally enough, since I'm a lexicographer by training, has favored America's so-called "college" dictionaries as a long-lived national institution with enough authority to command public trust and enough information to support a full-range measurement scale for definition-first vocabulary tests.So I recognize that others may well favor other targets:mathematics, perhaps, or cultural literacy, or even a comprehensive standard that includes the requirements set forth not long ago by the President's Council on Physical Fitness.

But dictionaries, I fell obliged to point out, touch us where we live and must live, namely, a social existence under the spell (and the spelling) of what is certifiably the most frustrating language on this planet, call it "English" (what on earth are "Engs"?) or "Latino-Germanic."Patriotism aside, our destructive Great Vowel Shift has isolated us from our literary past — Dante is far more intelligible to Italian teenagers than Chaucer is to American PhDs — and our Latinate technical vocabularies are far more accessible to the memorization cultures of Chinese, Arabic, and Hindi than to our most precocious high school graduates.

Bluntly put, our so-called English vocabulary is a Frankenstein's monster: a giant cobbled-together thing that rules us far more than we rule it, individually or collectively.Whatever standard measure of performance takes shape in response to the Pearson Educational Measurement white paper, I hope that there's room in it for a dictionary-based measurement scale.From first grade up to the first symptoms of senile dementia, something like this is what Americans need — all of us.