By Dan Marlin
Many of our students were finally able to take real, live ACT exams in September and October after so many sites understandably closed down in the spring and summer. We still have one more weekend of testing left in October, but we already have scores from September and some trickling in from the first October test administration.
You probably know how ACT scores work: English, Math, Reading, and Science section scores (each out of 36) are averaged to get a composite score of 36 on the full test. And scores round up – for example, a score of 29.5 is reported as a composite score of 30. But there are other intricacies you might be wondering about, so this post attempts to address those.
What is superscoring again?
Before the pandemic, the ACT had planned to roll out three new initiatives this fall: Computer-Based Testing, Section Retesting, and reporting superscores. Because priorities shifted to getting as many students into testing centers as possible, the first two of these initiatives were postponed for now, but superscores remained. Superscoring takes a student’s best score on each section from all of the official exams they’ve taken and averages the scores to create a new composite, which is often higher than the individual tests alone. Here’s an example:
|Section||Test A||Test B||Superscore|
In the past, colleges had to collect multiple tests to calculate superscores; now, the ACT will also report them. Many colleges will accept superscores, but some still do not, so it’s a good idea to check with your school’s counselor, or your Galin college counselor, before submitting scores to schools.
How does the writing section factor in?
It doesn’t! The writing section (scored on a separate scale of 2-12) has no bearing on a student’s composite score, no matter how the ACT may combine it with other scores on its reports.
How do raw scores convert to actual scale scores?
Indulge me while I get into the weeds. You can probably guess that the ACT scales vary from test to test, but perhaps not in the way you might expect. There is no “curve” on the ACT; a student’s score will not be affected by the other students who happen to be testing on the same day. Instead, the ACT makes small, statistically-based adjustments to each test’s scale to account for changes in difficulty, irrespective of the students who are actually taking it.
We can see an example of this by looking at the English section of two recent ACTs. The columns show how many problems a student can miss and still achieve each of the scale scores from 18-36:
|ScaleScore||Problems missed||ScaleScore||Problems missed||ScaleScore||Problems missed|
|Test 1||Test 2||Test 1||Test 2||Test 1||Test 2|
There’s not much variation at the upper end — for example, on the first test, a student had to be perfect to get a 36, while on the second test the student can get a “perfect” score while still missing one question. This small difference implies that the content on the second test is more difficult than the first, as its scale is more lenient. We can see this pattern more dramatically at lower scale scores: missing only 30 problems on the first test’s English section yields the same scale score as missing 34 problems on the second test’s.
Over time, I’ve noticed that scales have gotten ever so slightly more forgiving. This checks out — it appears that the math and science sections in particular have gotten a little more challenging as time goes along, so it’s important for the scales to keep up. That’s not to say that every section on every test will follow this trend, but in general, things stay pretty consistent.
If your student is doing ACT prep with us and is curious about the raw scores they would need to achieve a certain scale score, they can review the official ACT “red book,” which includes 5 practice tests and their scales. Your student should also discuss score targets with their tutor.
So why is it useful to understand how the scoring works?
I find it helpful for students to understand how raw scores translate to scale scores, for two reasons. First, it can help with strategy — students struggling with timing can still get a strong scale score even if they don’t complete every single problem. Second, I like showing students that, often, they can improve their score by just getting a handful of additional questions correct. For instance, a student with an English score of 26 on either of the tests above would need just two more right answers, out of 75 on the entire section, to improve to a score of 27. Confidence can be just as crucial as content and strategy as students seek to improve, and small score increases over the course of several tests can lead to big changes overall.
What about the SAT?
You had to ask, huh. The SAT is a bit more complicated with its scaling and curving, so maybe that will be the subject of a future post. For now, suffice it to say that students get an Evidence-Based Reading and Writing score out of 800 and a Math score out of 800, which are then added together for a maximum composite score of 1600. As with the ACT, if a student takes the writing section, that score will be separate and not factor into the composite.
The ACT isn’t required to release all of their test booklets, so unfortunately, we won’t know exactly what the scales were for the last month’s batch of tests. But by reviewing scales from other practice tests, students can get a good idea of how many questions they missed on official exams, and may find they only need to get a few more questions correct next time to achieve their goals.