Tackling the GED

I hope you’ve all had a little time to decompress after the holiday. We’re officially back in business now, though, so strap on your edu-gear and prepare yourselves for a beefy post. Today, my friends, we talk GED.

A couple of weeks ago, the Colorado State Board of Education voted to approve three separate high school equivalency exams: GED, HiSET, and the TASC. Holy acronyms, Batman!

Assuming the state can successfully negotiate contracts with the relevant vendors—GED Testing Service (now a joint venture of the American Council on Education and testing giant Pearson Education) for GED, Education Testing Services for HiSET, and McGraw-Hill/CTB for TASC—those looking for a high school equivalency diploma will be able to choose which of the three tests they’d like to take. Since the vote, I’ve had a number of people approach me about my thoughts on the idea of a “menu of tests” in the world of high school equivalency.

It has taken me a little time to fully collect my thoughts on the shift, but I think I’m there now.

The GED has been the high school equivalency test since it was created in 1942 by the American Council on Education, or ACE, a nonprofit organization made up of the presidents of “accredited, degree-granting institutions” of higher learning. In 2011, ACE teamed up with for-profit Pearson to “develop a new GED Test aligned with Common Core State Standards that will assure colleges, universities and prospective employers that adults who have passed the GED Tests are prepared to compete and succeed in a global economy.”

While the real reasons for this new “public-private partnership” are not immediately clear to me, it wasn’t long before I started seeing the argument that this is yet another example of an evil, profitizing company maliciously seeking to grow its profit margins on the backs of corporately enslaved children.

I find this argument somewhat confusing coming from people who advocate for the other two tests. The TASC test belongs to McGraw-Hill, the only other education company that could plausibly be said to rival Pearson. Similarly, HiSET’s Education Testing Solutions, though a nonprofit, maintains for-profit subsidiaries as allowed under the law, and has been accused of masquerading as a nonprofit while watching its coffers swell. Fighting back against a for-profit company by bringing in other organizations that profit on education seems a little… off key.

Regardless of all that, though, this line of argument doesn’t do it for me. I will admit that I often find myself uncomfortable with Pearson’s apparent attempt to build a monopoly in the education space. However, I continue to be unconvinced by general arguments against companies “profitizing” education. Unless we are prepared to forgo pencils, paper, desks, chalkboards, computers, buildings, food, backpacks, and crayons, or have the government produce those items for us, we should probably abandon the notion that profiting off of anything related to education is immoral and focus simply on who can provide the better product or service.

In any case, why the partnership happened is much less important than the new GED it produced. The new test launched in 2014, and a flurry of angst quickly followed. A reporter in Cleveland chronicled a massive drop in the number of people taking and passing the test, the Washington Post ran a similar story, and a totally weird hodgepodge of political spectators from the conservative Daily Caller to ultra-left reform opponent Diane Ravitch also piled on. Here in Colorado, the progressive Bell Policy Center jumped into the fray with a lengthy report asking the State Board to provide more high school equivalency exam options.

Some of the criticisms leveled against the GED have undeniable merit. Certainly, we can’t deny the fact that the number of GED test takers plummeted in Colorado when the new test was introduced. Official national 2014 numbers are not yet available, and I’m leery of “estimations” when it comes to this type of data. But the Bell policy report linked above does include Colorado-specific data showing a drop in the number of test takers and test completers.

In 2013, 18,714 people took the old GED in Colorado—the highest number since at least the turn of the century. That bump was likely spurred by the news of a harder, more expensive test coming down the pike. In 2014, by comparison, the number of test takers was 4,469. That’s a roughly 76 percent drop. (Notably, this trend may have bottomed out and started to reverse; 5,573 people had taken the test in 2015 as of October.) Similarly, the percentage of testers who actually completed the GED fell from 77 percent to 54 percent.

What happened? Why did the numbers plummet so dramatically? What’s wrong with the new GED? Plausible arguments include:

  • The cost is too high. In Colorado, the new GED costs $150 for the full test, or $37.50 per module. Those amounts must be paid by test takers themselves, and are significantly higher than they were several years ago—though we should note that costs began increasing before the advent of the 2014 GED.
  • The test has to be administered on a computer. This could limit the number of testing centers, many of which are private organizations (and some of which are, by the way, for-profit companies) that simply contract with test providers to administer certain tests, able to accommodate the administration of the test. Also, some people just don’t like doing stuff on computers.
  • The test is too dang hard! As evidence of this excessive hardness, folks cite the fact that the number of Colorado test completers who passed the 2014 GED was 70.5 percent—a 15 percentage point drop from 2013’s 85.7 percent pass rate.
  • The test has been contaminated by Common Core. This is true enough, but only insofar as it aligned itself to the new standards all at once instead of doing so gradually as the other two tests are doing. According to the Bell report, all three tests will be fully aligned with Common Core by 2017, effectively rendering the related arguments for or against certain tests moot. Changing the standards is the only way through this particular Gordian knot, but that’s a different conversation. I’m not going to belabor the point any further.

The first point about cost is definitely a valid one. Costs were trending upward before the new test came out, but the fact remains that it spurred a sizeable increase in cost. The new GED is an undeniably expensive test. Then again, GED’s vendor fees are only $80 for the full test, and $20 for individual modules. The remaining costs are due to state fees for… gobbledygook that I won’t pretend to know about.

Still, though, HiSET’s and TASC’s vendor fees are much, much lower at just $50 and $52 respectively. Obviously, this makes them far more accessible for a wider group of people. And given the fact that GED test takers tend to come from socioeconomic backgrounds in which even small increases in costs can be problematic, the increased cost could indeed help partially explain the lower numbers of people taking the test. It also raises important concerns about whether the test is well positioned to do what it is intended to do: provide a path to a better life for folks who never finished high school for one reason or another. We’ll count this one as a strike against the GED.

The second point is less damning. I have no doubt that some testing centers will find themselves unable to administer the new test due to a lack of equipment or expertise. But technology is the future, and it provides a path around the cumbersome processes associated with collecting, securing, transferring, and grading hard-copy standardized exams—all of which should have the effect of driving down costs over time. It also allows for fast turnaround time with scores (unless you’re talking about PARCC, that is). It’s hard for me to believe that old-school testing centers will not need to adapt to computer-based tests in the coming years, no matter how much we try to forestall that inevitability.

The argument that “some people just hate computers” is similarly unconvincing. We live in a world where digital literacy can mean the difference between success and failure, and where not knowing how to operate a computer is likely to preclude you from filing your taxes, paying your bills, or even working at McDonald’s. And unlike the K-12 space, where computer-based requirements can cause massive headaches due to schools needing to test every child in a certain period of time, high school equivalency tests are fully optional and administered at dedicated testing centers. So no, I don’t think it is unreasonable for us to expect adults taking a high school equivalency test to do so on a computer. In fact, I think it may be a good thing. (Funnily enough, the Colorado Department of Corrections agrees with me on this point.)

The third point, which deals with difficulty, is where things get really interesting. It is true enough that 2014 saw both a decrease in the number of GED test takers and a decrease in pass rates. But according to CDE data covering fiscal year 2014-15, 77 percent of all Colorado GED completers passed the test between July 2014 and June 2015. According to the Bell Policy report, 2015’s pass rate among Colorado test completers was up to 80.5 percent by October—just two percentage points shy of Colorado’s average pass rate (82.7 percent) over the five years prior to the introduction of the new test.

When you look at national data for HiSET and TASC (there is no Colorado-specific data for these tests for obvious reasons), the “too hard” argument falters further. HiSET’s national pass rate in 2014 was 75.1 percent. That is significantly higher than GED’s 60.9 percent national pass rate in 2014, but not shockingly higher than the 70.3 percent GED pass rate as of August 2015. Meanwhile, TASC’s 2014 pass rate was a dismal 58.9 percent, and that rate looked to be even lower (57.6 percent) as of August 2015. (See CDE’s helpful chart for citations of the above information.)

We should acknowledge that TASC’s ugly overall pass rate in 2014 was significantly dragged down by New York, where only 49 percent of the state’s roughly 24,000 test completers passed the test. In Indiana, the state with the next largest group of completers (about 7,100), the pass rate was 79 percent. In West Virginia, it was 81 percent. It looks very much like TASC’s low national pass rate is thanks to a struggling adult education system in New York. If you remove that outlier, the test’s pass rate lands in the same neighborhood as HiSET’s, and somewhat higher than GED’s.

In fairness, I can’t draw real, mathematical comparisons between different tests’ pass rates, or between Colorado and national data. There’s a lot of nuance involved in the concept of test difficulty. Pass rates are only one proxy for that concept, and they are a proxy based upon the assumption that all three test makers are using sound norming and standard-setting practices in roughly similar populations.

But as long as we accept that assumption—and it seems like a relatively safe assumption to accept given the pedigree of the organizations involved—I honestly don’t see a tremendous difference between the tests when it comes to difficulty as evidenced by pass rates. The GED looks to be somewhat harder overall right now, but if GED’s upward trend continues, or HiSET and TASC see pass rates fall as they finish aligning with Common Core, all three tests may well wind up with near-identical pass rates. (Note that these rates do not account for increasing numbers of people who are not finishing the GED–a potentially serious issue with too many possible causes to be adequately tackled here.)

That’s not the end of the story, though. CDE says that even if all three tests are offered in the state, Colorado’s institutions of higher learning still get to decide which credential they’ll accept as evidence of college readiness. The same is obviously true for employers. Some members of both groups may not accept a credential awarded by a test they see as less rigorous than the others, which means that there is a chance some of the folks high school equivalency tests are designed to help may find themselves locked out of opportunities simply because they chose (and paid for) the wrong option. That’s why it will be important for psychometricians (i.e., not Little Eddie) to make sure the tests are all comparable to one another—a very tough thing to do in a “menu” system. I would be very interested to know how other states offering more than one high school equivalency exam have fared statistically under their menu systems, though I have yet to see any compelling research on that topic–and research on pseudo-menu systems in the K-12 world is… unfavorable, to put it gently.

Comparability is important because our end goal is to make sure people who earn a high school equivalency diploma are prepared for the challenges that await them afterward, regardless of whether those challenges are in the work force or at college. Our end goal is not—and should never be—to artificially beef up numbers on a spreadsheet by fiddling with tests. As is the case with traditional high school graduation, whether or not people are actually prepared for the next phase in their lives is an awful lot more important than how many slips of paper we issue.

So where does this leave us? I suppose the answer is still a bit of a question mark. If all three tests can be proven to be comparable to one another and rigorous enough to satisfy our state’s current requirements for a high school diploma, I see no reason not to allow people to choose which one they’d like to take. That’s especially true given the fact that the new GED is clearly locking at least some low-income people out due to its high cost.

However, if the tests are not comparable—meaning that one is easier than the others, or tests different things, or otherwise fiddles with stuff, etc.—the better course would be to reject the “menu system” and instead advocate for a single rigorous test (and no, that wouldn’t necessarily need to be the GED) that can meet our state’s needs.

Why? Because I believe that simply providing an easier path for people is no different than implicitly deeming them incapable of meeting tougher expectations. Or, put another way, succumbing to the “soft bigotry of low expectations.” Maybe that makes us look good on our spreadsheets, and maybe it even helps some people in the short run, but it’s hardly a kindness in the long run. Colorado secondary education—traditional or adult—should be in the business of pushing students to achieve more in an increasingly competitive world, not lowering the bar until we can get enough people across the finish line that we all feel okay about ourselves. We ought to be in the business of handing out well-earned tickets to future opportunities and success, not participation trophies.

If we can do those things with a menu system, so be it. If not, we’ll need to adjust our course. In the meantime, Little Eddie will be watching.