Yay For College App Test Out Options – But Read The Fine Print

Look at your social media feed today and you may have felt a wee bit of relief that yet another college announced SAT and ACT scores are no longer required for undergraduate admissions. For test-weary students, knowing 800+ colleges offer some type of standardized testing opt-out sounds like good news. But, before you recycle the test prep manuals, be sure to read the fine print.

Beginning August 1, 2015, prospective freshmen interested in applying to George Washington University have the option of skipping the SAT and ACT tests entirely. That is, of course, unless they are homeschoolers – OR – they come from a high school that didn’t issue traditional grades – OR – they’re recruited NCAA Division I athletes – OR – they are applying to a special 7-year accelerated BA/MD program. The Admission’s website explains: “The best indication of whether a student will be successful at GW is their performance in high school – the grades they earn and the rigor of their coursework.”

Excuse me while I scratch my head for a moment. GW will exempt SAT and ACT test scores for a super strong GPA? That new admissions’ policy doesn’t make a whole lot of sense to me, knowing that inflated high school grades have lead to 30% of college freshmen dropping out of school in their first year.

And let’ not forget about cheating. Not long ago, Education Week reported how one-third to one-half of classroom students have unabashedly admitted to cheating. Whether it’s buying a paper that someone else has written for them or looking up answers on their smart phone during a test, traditional classroom students are literally doing whatever it takes to make the grade.

The good news is, other testing opt-out colleges are trying to be more equitable in their admission process.

Brandeis University in Massachusetts, for example, offers freshmen applicants an opt-out testing choice. Choice A (which is not really much of a choice) allows students to substitute a combination of scores from three SAT-Subject Tests; APs; or IB exams instead of traditional SAT or ACT scores. With Choice B (which is a pretty cool choice), applicants can submit a Portfolio that includes “one graded analytical writing sample from 11th or 12th grade”. That is, of course, unless you’re a homeschooler.

Temple University in Phildelphia took an entirely different tack, offering the Temple Option, where prospective students “submit self-reflective, short-answers to a few specially designed, open-ended questions instead of their SAT or ACT scores”. Unless, you guessed it, you’re a homeschooler.


Having conducted my fair share of end-of-year homeschool compliance reviews, I completely understand the need to assess homeschool students by something more than parent-issued grades when you’re applying for post-secondary schools and jobs. Just as you find in any school choice population, a few unsavory types will try to pull a fast one, in terms of proving accountability. All the same, I think these college testing opt-out policies unfairly target homeschoolers and require a separate and unequal set of admissions standards.

I’m not going to point to studies suggesting that homeschoolers out-perform public school peers on standardized tests and, therefore, should be automatically assumed to be qualified to enter any college of their choice. Frankly, I think those studies are flawed by their self-selection bias and make for a weak argument. But, to automatically assume that mommy grades on a homeschooler’s transcript are automatically inferior or suspect is both offensive and naive.

If colleges suspect the validity of grades for some students, then what about online schools? The National Education Policy Center found online K – 12 students being educated at unacceptably low levels. Yet, these students are still passing classes, in part because they can take an exam over and over again until they pass it. Add to that the fact that some online schools don’t require face-to-face interaction to verify which student is actually completing the work. Still, college admissions officials do not seem to feel cyber students need to be held to a different level of testing opt-out accountability – in the same way as homeschoolers.

I applaud the proactive step some colleges are taking to help end the testing frenzy we find ourselves in as a country. Clearly, alternative methods exist in selecting students who can succeed in college and I look forward to seeing future data that tracks the opt-out testers over time.

My hope, however, is that colleges rethink how they view homeschoolers in this process and create a more equitable testing option process.


What do you think of the double standard testing option policies?
Share your thoughts below!




Google+Google BookmarksGoogle GmailLiveJournalLinkedInEvernoteShare

Grade Equivalent Score Fallacy

What Do Grade Equivalent Scores Really Mean

Grade Equivalent scores remain one of the most misunderstood and misused pieces of data from educational testing. Rarely explained in a test report, parents who see “GE > 18.0” sometimes come away thinking their child is achieving at a post-graduate school level.

While a child who obtains such high GE scores is intelligent, no doubt, it is unlikely they are ready to jump into graduate level courses on their own – especially if they’re only 12 years old.

The National Association of School Psychologists explains Age and Grade Equivalent scores simply:

[I]f Jacob’s performance on the test of reading comprehension is equal to an age equivalent of 8.7 years and a grade equivalent of 2.6, this means that his obtained raw score is equivalent to the same number of items correct that is average for all 8-year, 7-month old children included in the norm group on that particular reading comprehension test.

Let’s take a look at a couple of hypothetical test profiles to better understand Grade Equivalent scores.

Grade Equivalent Score Comparison Chart

Before I talk about these numbers in detail, let me give you a quick primer on some of the terminology.

Standardized tests do not report scores as simple percentages of right answers. Instead, a student’s Raw Score (the number of questions that were answered correctly on each subtest) is transformed into a Standard Score. A Standard Score of 100 is considered Average. Standard Scores can run high or low, in either direction.

If you want to compare a score to other test takers of the same age or grade, you can look at the Percentile Rank. For example, a Standard Score of 100 equates to a 50th Percentile Rank. Put another way, if you lined 100 kids up with highest to lowest Standard Scores, someone with a 100 would be right in the middle of the line – with half the students achieving less than them and the other half achieving more.

Grade Equivalent scores, on the other hand, allow us to compare the total number of correct answers the average test taker got. For example, an average 12-year old taking the 3 subtests that make up the Broad Math portion of the Woodcock Johnson-III Test of Achievement would need to get a total of 141 correct answers out of a total of 268 possible questions to score at the 50th Percentile. How that test taker got those 141 correct math answers will depend upon the individual, but more than likely, the average person got some questions wrong as they worked their way through the test.

So, back to the chart. Mary is our typical 12-year old girl. When we look at her math achievement scores we can feel rather confident that she’s achieving at grade level and is most likely doing well in her actual schoolwork.

Abby, on the other hand, is 7-years old, the typical age for a 2nd grader. Her achievement scores suggest she’s a pretty smart cookie. Some people may even take a look at her Grade Equivalent scores and think she’s a candidate for radical acceleration – but that’s not necessarily the case.

With Percentile Ranks of 99.9, we can say confidently that Abby is achieving, hands down, beyond expectations for 2nd grade. Abby would be an excellent candidate for gifted programming and probably even at least one grade skip.

However, just because her GE is in the 6th grade range in Calculations and Applied Problems, it does not mean she’s achieving at the 6th grade level – or even that she’s capable of accomplishing 6th grade level work, as of today.

Let me explain why.







The WJ-III, just like the Wechsler Individualized Achievement Test (WIAT-III), samples a person’s level of achievement across broad content areas. Neither test was designed to exhaustively assess if a person has learned all the goals, objectives, and content usually taught at each grade level. Tests like the Terra Nova and the Iowa Test of Basic Skills do that.

Take for example the Calculations subtest on the WJ-III. On this subtest, students have the opportunity to answer up to 45 math questions that range from basic addition to calculus. However, 22 of those questions focus on the four basic operations using only whole numbers. Only 2 questions assess calculus knowledge.

In the case of our hypothetical 7-year old Abby who answered 21 questions correctly on the Calculations subtests and earned a GE of 6.2, it’s possible that she aced all the basic operation problems, but failed to get one fraction or decimal problem correct, let alone even attempt a pre-algebra or higher math question. That doesn’t take away the fact that Abby is clearly ahead in math, but, at the same time, it doesn’t make the strongest case for saying that she should be in 6th grade math.


Grade Equivalent scores can be used to compare the number of correct answers children of different ages or grades received on the same test. Those Raw Scores, however, will lead to different Standard Scores based upon the test taker’s actual age or grade. Grade Equivalent scores do not tell us that a child is actually achieving at a specific grade level.

 

A high GE score tells us that a child has been able to correctly answer far more questions than his or her peers – but it tells us nothing more. At the same time, a high GE allows us to infer that the student more than likely has the ability to handle a greater breadth or depth of material than they are currently encountering, if they are in a typical age-grade placement.

Just how advanced the material should be is a question better judged by examining work samples and talking directly to the child. If you are attempting to advocate for a grade skip through a school, requesting that your child take the end-of-year assessment test for a specific grade level subject will provide you with stronger data.


Alessa Giampaolo Keener, M.Ed. works as an educational diagnostician in Maryland. She holds a Masters degree in Education from Johns Hopkins University and is recognized as an expert witness on matters of educational best fit for special student populations. Alessa also works with families and schools to better understand test data and how that information can help educational planning.


What’s been your experience with Grade Equivalent scores?
Share your story below!