The fiscaltimes finances

THURSDAY, MAY 3, 2012

Part 3—Doesn’t know, doesn’t much seem to care: Is Pearson over-charging the state of Texas for its testing services?

Rather clearly, this was the “central point” of Gail Collins’ column in Saturday’s New York Times. Right from its headline on down, Collins seemed to say that Pearson’s “monster profits” were a rip-off—that the company is massively overcharging for its “pricey” efforts.

But how strange! At no point did Collins make a clear statement of this alleged “central point.” What was she claiming in her column? In truth, it’s quite hard to say.

Collins pleased liberal readers with her many complaints about “the privatization of education.” But she never defined her alleged “central point.” In particular, she showed no sign of knowing whether Pearson’s profits involve an actual rip-off.

Can we talk? Collins showed no sign of having tried to examine this question at all. She showed no sign of having tried to do any actual journalism. As he typed her scattershot piece, Collins expounded on a large array of familiar educational topics. But thanks to this cynical game of hopscotch, she didn’t have to expound at length on any particular issue.

She complained about for-profit textbooks; she complained about for-profit tests. She complained about charter schools. She complained about schools which are “completely online, with kids getting their lessons at home via computer.”

She complained about for-profit after-school tutoring, echoing the strangest thing Diane Ravitch ever said in this area. (More on that complaint tomorrow.) She complained about a possible change in the GED program.

Eventually, killing a bit more time, our slacker complained about this:

COLLINS (4/29/12): The Obama administration has been trying to tackle the astronomical costs of 50 different sets of standardized tests by funding efforts by states to develop shared models—a process you will be stunned to hear is being denounced by conservatives like Gov. Rick Perry of Texas as “a federal takeover of public schools.”

Almost surely, Collins had a good point here—although there was no sign that she actually knew this.

Let’s talk about “the astronomical costs of 50 different sets of standardized tests.” As a bit of background, let’s consider the way standardized testing was done in and around the year 1970.

In those days, annual testing was already widespread, but it involved a set of “norm-referenced” tests produced by roughly half a dozen major test publishers. Example: The Iowa Tests of Basic Skills, a battery published by Houghton Mifflin.

About a half-dozen such batteries were in widespread use around the nation. As a general rule, new versions of these tests were produced every seven years.

There were shortcomings to this seven-year itch, especially when some school districts began to cheat on these tests, bowing to the new age of “accountability.” (In our experience, this overt cheating was already under way in the early 1970s.) But with a half dozen major test batteries being revised every seven years, test publishers could devote a great deal of time and effort to the creation of the new batteries.

We'll take a guess: Almost surely, those batteries were of higher technical quality than many of the statewide tests which are in use today. Here’s why:

We now have fifty different states producing fifty different test batteries. Often, the tests are changed on an annual basis. Back in 2006, Collins’ own paper, the New York Times, published a news report about the problems being caused by the sheer volume of these fifty different test batteries.

Uh-oh! With so much test development occurring, the nation was facing a shortage of “psychometricians,” David Herszenhorn reported:

HERSZENHORN (5/5/06): With federal law requiring wider testing of schoolchildren, the nation faces a critical shortage of people...with the mathematical, scientific, psychological and educational skills to create tests and analyze the results. The problem has sent states, testing companies and big school districts into a heated hiring competition, with test companies offering salaries as high as $200,000 a year or more plus perks.

A result is a peculiar outcome of the No Child Left Behind act. Psychometrics, one of the most obscure, esoteric and cerebral professions in America, is also one of the hottest.

These experts are needed in virtually every aspect of developing, administering and scoring exams, from deciding what test will best measure certain skills to drawing up questions and answer sheets. Doctoral programs are producing at most 50 graduates a year in the field.

"This was always a very, very tight, small group of individuals prior to No Child Left Behind," said Wayne J. Camara, vice president for research and psychometrics at the College Board, which publishes the SAT and Advanced Placement exams. "Since No Child Left Behind, it has just gotten ridiculous.”

[...]

Government and industry officials warn that the shortage of experts could undermine the testing process and lead to errors, with consequences like children's being wrongly denied promotion and schools being mistakenly labeled as failing.

Already, they say, many states and school districts lack officials trained to oversee testing and make effective use of score data. The states are being hardest hit because they desperately need psychometricians to supervise their multimillion-dollar contracts with test publishers but are routinely outbid not just by testing firms but also by colleges, research groups and other industries.

Does this country have enough “psychometricians” to produce the current level of tests? We often wonder when we read about various breakdowns within the various states’ testing programs. We wondered again when we read Kevin Drum’s treatment of the minor flap which led to Collins’ rambling, know-nothing column last week. (For Drum’s treatment, click here.)

Collins may have brushed up against an actual problem when she cited “the astronomical costs of 50 different sets of standardized tests.” It does cost a huge amount of money to produce fifty different sets of tests. It also may require more expertise than we have available.

That said, there is no sign that Collins knew what she was talking about when she penned her single sentence about this alleged problem—a sentence which led to a pleasing jibe aimed at the vile Rick Perry. But then, we saw no sign that she knew squat or squadoosh about any of the various problems she pretended to discuss in this worthless but pleasing column.

Back to Collins’ “central point:” Is Pearson over-charging Texas for its testing services? Is Pearson over-charging New York? Is something wrong with this company’s alleged “monster profits?”

Should those states just produce their own tests?

There was no sign that Collins knew squadoosh about these questions, or about any of the long string of topics she ran through this day. But she had produced a column which her readers found highly pleasing.

Tomorrow, we’ll review their clueless comments. And we’ll see why we liberals fail.

Tomorrow: Ever so slowly, the truth begins to emerge