A testing paradigm for the 21st century

Okay, let’s begin with some points of agreement: 1) Common Core has been a disaster; 2) No Child Left Behind is better labeled as “No Child Allowed to Learn”; 3) the mania for testing children repeatedly, with no chance to recover from a bad day or a misread question, is doing very little to improve the educational system in the United States.

I also have a streak of realism in me, and know that this goal, if not entirely infeasible, is at very least going to require a complete shift in public awareness, if not an entire quantum leap in the evolution of the species. In the meantime, therefore, I keep coming up with measures to make stepwise improvements, so long as they do not backslide along the way from the road to human Liberty. This might be one of them.

Still with me? Good! Now let’s take a look at how this might be done better, for the benefit of students, teachers, and even those educrats who think they know, more than anyone else, about (to give them the benefit of the doubt) helping children to truly learn—not what to think, but how to develop their own abilities to distinguish truth from nonsense or outright lies, and to form their own views about life. (I will grant that some of those in this field are less interested in seeing this happen than in creating another generation of Prussian factory-drones; they aren’t reading this anyway. This is addressed to you folks who actually do care about literacy, numeracy . . . and the facilitation of student learning.)

Full disclosure: For the past 15-plus years, I’ve worked with student testing, in one form or the other. A lot of this has been as a Scorer for a wide variety of those standardized tests being taken, from all across the country, to meet state (and now federal) requirements for every student, whether said student be in a government-run (aka “public”) school, a private one, or under the broad spectrum of homeschooling, unschooling or other autonomous systems of learning. In that capacity, I have scored writing essays, math and science word-problems, and even a few Field Tests for questions being considered for formal inclusion. (Note: In the latter case, the poor guinea pigs … err, students often don’t even know their answers will not affect their evaluations in the least—or else they tell them, so that over 70 percent don’t even bother to answer the questions.)

I more recently supplemented this for a time by scoring tests for an online school, providing a curriculum for high school students in the region, offering both make-up and advanced-placement courses (as well as, I’m happy to say. some “life skills” ones to aid students in moving on with their actual lives after the formal schooling ends). With this seeming vested interest in expanding the testing game, my proposal might look like a means to doubling my workload, by expanding the testing season; actually, I’m seeking ways to shift my time to more valuable purposes.

The system as it is: As it now stands the process is the same in almost every school. Students are tested, at least once a year, from about third grade to junior year (I refuse to say “11th grade”), in a rather grueling stretch of several days devoted solely to the tests. In most schools, this is preceded by a single-minded focus in each classroom, covering most of the time before the test period, on: a) how to take the tests; b) what kinds of questions might appear on them; c) what specific skills are being tested for (about which some teachers might have a clue, though most find it just as bewildering as the students do); and d) whatever other agendas have been dictated this time around. (Many teachers I know are irate at their students being made guinea pigs or beta-testers for whatever ivory-tower teaching or learning theories have come into vogue in each round of this process. The Common Core “new math” methods are just the most obvious example of this atrocity.)

The students then take the tests. Along with the multiple-guess questions scored by machine, there are numerous essays and short-answer questions requiring human eyes (and minds) to evaluate properly. Although there are strict rubrics (as well as “range-found” papers against which to compare them) for assessing the proper scores, a considerable percentage are still deemed unscorable under the present paradigm. There are several reasons for this:

Some students completely misunderstood the question. (Although this happens more frequently when said student is laboring under the handicap of English being a second, third or even fourth language, it also hits American-born students lacking basic awareness or training in what is supposed to be their “native language” as a primary means of communication.) These unfortunates might grab onto one word in the question and run with that, perhaps thinking that by giving a good definition of that one word, they’ll get enough credit to scrape by;

Some students either don’t know the answer, don’t really care to do the work to find it, or lack the skills to express themselves with even marginal coherence (some draw pictures, some enter random keystrokes, some go on a rant about how stupid the tests are); and

Some students, now that the tests are being given online, are under the impression that merely copying the right piece(s) from the existing text is sufficient for a score. This was once confined to copying from the prompt, but now extends to wholesale copy/paste from the text of a reference essay, or even a website, when the assignment asks specifically for an essay written in the student’s own words.

Each of these types of papers might merit some kind of score; more often than not, they all go into the category of “Unscorable” and are then returned to the schools, showing that for the question, which is deemed only slightly better than no attempt to answer it. With more than a few of those Unscorables, along with a few ZEROS or blanks, the student shows up as a failure on the test. (As there is no way to evaluate the work, there’s also no way to give credit for it.)

I fear these students are being lost between the cracks, just as at one time some very intelligent young folks (some bearing a true learning disability or perceived “behavioral problem,” others who just did not fit the mold) got stuck in the “dummy track”; without outside intervention to restore self-esteem and a sense of worth, many of them (including some potential geniuses) were set aside, given “busy work” and advised to expect driving a truck or digging a ditch as their highest aspirations in life. (I’m not saying there’s anything wrong with driving trucks or digging ditches, only that as a lifelong career it’s a bit limiting, if there’s an architect, lawyer or brilliant scientist hiding inside you!)

I actually know of several such cases, including one who spent most of his high school days (early 1980s, in a major metropolitan government-school system) sitting at a desk with an open dictionary. His assignment was to copy a full page of that dictionary, and when he had finished he could lay his head on the desk and nap for the rest of the day. He was eventually encouraged to seek more, and jumped the system to get an associate’s degree in real estate law from Harvard University. He then went on to become a successful realtor running his own business in the late ’80s (until the industry went belly-up a few years later).

As I said earlier, I’ve seen the effects of all this, initially at a distance, helping my friend—who was indeed almost fatally dyslexic—to write his papers for school. (He would pace the room and dictate, as I typed and cleaned up the syntax, while learning a bit about easements and property law in the process.) More recently, over the last decade and a half, I’ve encountered the issue while evaluating some of these papers.

In my former posting with the online high school courses, I also saw what might be a way out. One thing about virtual work is that the scorers of the questions are working with an interactive system, and so can offer feedback; more importantly, they can also give second (often, even third) chances to students, whether deserving or just clearly bewildered. (I’m not certain, but I might have been the first scorer for these papers to pick up on this latter process.)
When I read one of these essays (sometimes just one or two paragraphs), I look for four things:

a) Are these the student’s own words? Since the material is being researched on the internet, this is easily determined, using a Copyscape.com account, which can smoke out a text-copy at 100 yards. If it’s not original work, I attach a note about text-copying, score it a ZERO, and click the “Retry” button.

b) Does it cover the question sufficiently? If so, it gets full credit, as well as either a “complete answer” or “excellent essay” notation, depending on the overall quality. (Sometimes they’re so well written I’m compelled to check the Copyscape stats; if no copying is then found, I give that student proper encouragement by noting this, then thanking him or her for taking the course so seriously.) If it’s seriously deficient in either detail or grammar/syntax/etc., I generally offer the student another chance to fill it out a bit more (or clean up the sloppy writing) for a better score.

c) Is it on topic, and within the bounds of the question and the course? (A History paper should include dates and places, not just be a general ramble about the purported subject; a math paper needs to show the work, enough so it’s clear the student grasps the concept, knows how to use the formula, etc.)

d) Does this student understand the question, even if (due to language difficulties, etc.) (s)he seems to be way off base?
Case in point: I had one student about a year ago who (from the syntax issues) was probably coming from an Asian heritage and vernacular; it took me three readings to realize that she fully understood the math problem and had given a valid answer, but lacked the vocabulary to bring it down to simple terms—every complex polysyllable was a synonym for the word she was searching for! I gave her almost full credit, with a point or two off for the syntax and usage issues.

Okay then, given these barriers and disconnects, what’s the solution?

Here’s a thought: Why not give these tests COLD, in the first month of each school-year, with little or no preparation, sending them to the companies to score them, and getting on with teaching whatever you should be teaching for that school semester? They would then measure whatever learning had seeped in, and stayed there, in those young minds, without the artificial effects of zero-hour cramming and force-feeding that neither teachers nor students either like or need. (If the intention is to measure what a student has actually learned, last-minute sessions of memorization and route drills have no valid role to play.)

Then, six months or so later, the tests would come back to the schools and the students, with feedback on their performances from the scorers, so the teachers could now focus primarily on those who need help. You would then give them the exact same test to do over again, as or if needed!

For students who need help—understanding the questions, writing their answers coherently, or otherwise—the teacher could conduct sessions, group or even individual, to clarify problems and answer questions, perhaps based on a report summarizing the students’ issues, received prior to this retesting period (which should be much shorter, for all but a few students). Although this might be the “weakest link” as far as ‘security issues” are concerned, it’s also where actual learning might take place; somewhere along the line we must restore some measure of trust toward our teachers, and let them do the jobs they thought they’d be doing when they chose the teaching of children as their careers.

For students who’ve already done well, and clearly do not need to go through the whole process again, there’s a much easier task. At their discretion, they might choose either or both of the following: a) spend some time reviewing any questions they missed, rewriting those they wished to improve their scores; or b) engage in some other kind of learning, perhaps an independent study project on some topic they really do choose to pursue. The computers are there, anyway (assuming the site-blocking is set up to somewhat limit access from just surfing for porn?); they might as well be used for something productive, something that encourages a student to pursue new learning.

This is all now quite possible, since the tests are being administered virtually, over the internet. With rare exception, students must now type their answers on a computer, creating an electronic record of their work. It could also be a perfect opportunity to receive feedback on the work—or even a second chance, whereby a paper originally written on a “bad hair day” or a question misunderstood due to ambiguous verbiage (you’d be amazed at how often that happens!) could be redone by the student, and submitted as a better representation of that student’s knowledge of the subject. (Currently, as noted above, these are just marked as “unscorable”; if enough students clearly misread the question, the rubric must be altered to cover an answer defying the educational intent of the question, so as not to penalize those students who saw something different.)

In the case of the student who had used the “right click-Copy/Paste” method to grab text from a reference document, that “second chance” might be the difference between a non-scorable paper and a top score on one. Although there is a tendency to penalize a student for doing this, as I have learned in my travels, it can be a cultural thing: For example, in many Asian and some Middle Eastern traditions, it is honoring someone’s wisdom to just repeat the words—even without proper attribution. For students coming these traditions, as well as others similarly unclear on the concept, it makes little sense to penalize them for these actions; a “second chance” (along with a feedback note: “You’ve done the hard work by finding the right answer, now write it in your own words!”) could give them an opportunity to turn that Unscorable into a successful paper.

All in all, this seems like a much more sensible way to evaluate student learning when using standardized tests. The technology is there; it would only involve a few tweaks to the software to add the Feedback and Retry buttons to the systems. The alleged goal of actually evaluating and promoting progress, instead of just penalizing failures (real or imagined), might also be achieved, and much more readily. Meanwhile, those weeks, if not months, now being devoted to “teaching to the tests” would become unnecessary, and the time could be refocused on actually teaching children some basic skills, leading them to love learning, and to learn for themselves about their world, their cultures and the rich heritage of freedom, responsibility, and community.