Whenever I'm working through a standardized test, I'm on the lookout for model questions—those that best encapsulate the feel, content, and challenges of a particular section. I'll often use these questions with students in order to demonstrate specific strategies and their applications. Once students master the approach for handling a model question, they unlock their potential to correctly answer a wide variety of similar questions, and thus to raise their scores significantly.

Recognizing Model Questions

Of all sections, the ACT Science is a frequent culprit behind stubbornly static test scores. Contrary to common assumption, the Science section score depends much less on raw science fluency than on the ability to make inferences, recognize patterns, and synthesize information from multiple data sources. It turns out that with a little model-based training, students can learn to associate different question types with specific, predictable procedures; when they successfully apply and execute these procedures, they avoid common traps, move through the section more efficiently, and answer more questions correctly.

An Example

What might a model Science question look like? Let's see an example from the December 2014 ACT. (Try answering the question yourself before reading the breakdown below it!)

A "suppose" question serves as a perfect model: this question type—which asks students to consider a hypothetical addendum to the experiments already outlined by the passage—appears frequently, demands specific, predictable types of critical thinking, and presents plenty of room for faulty assumptions. The proper chain of reasoning necessary for answering this question correctly is as follows.

1. The problem references Figure 2, at H = 4.0 m on the x-axis. I should go there and focus on the points on that vertical line.a

2. Spring W isn't on the graph, so I need to determine where it would be located if its curve were graphed.b

3. The text above the graph states that X is the least stiff spring and Z is the stiffest, with Y in the middle. Associating these facts with the position of the curves, I can conclude that the stiffer the spring, the higher its curve's position on the graph. Therefore, Spring W's curve would be above the others at all points where H is greater than zero.c

4. The correct answer is choice D, because it's the only y-axis value greater than those shown for the other springs at H = 4.0 m.

1. The first step seems like the simplest—go where the question tells you—but surprisingly, it cannot be taken for granted. Students lose a startlingly high number of points on the ACT Science simply because they look at the wrong figure. The solution? Habit formation: students need to be taught that when they see a reference to a figure or table, they must reflexively stop reading the question and make sure they go to the proper figure immediately. Only after they're in the right place should they continue reading the question.

2. New information (i.e., mention of Spring W, which wasn't part of the original experiment) requires new reasoning in the context of the data already presented by the graph. The need to determine how Spring W relates to the others doesn't always occur intuitively. Students need to learn that in this scenario, they must depend on a pattern or trend (evidenced in the graph or perhaps elsewhere) to gain insight.

3. Students often expect that when they're referred to a figure or table, all the information they'll need will be in that figure or table. This is incorrect, and often leads to faulty assumptions. For example, students who expect that Figure 2 itself will contain everything they need might assume that the curves just appear in alphabetical order; that is, they'll assume that because 'W' precedes 'X' in the alphabet, and because the curves seem to be arranged on the graph in increasing alphabetic order from low to high, Spring W's curve should appear as the lowest curve on the graph. As shown by the proper reasoning in step 3 above, this assumption is inconsistent with the facts. What's the fix? Students must become extremely familiar with the way questions like this go: in "suppose" questions, the ACT will often require them to integrate information from the text accompanying a figure or table. Only after they do will the proper relationships become obvious.

This question epitomizes the "suppose" model. When students see suppose, they should realize and execute on the following

1. Determine which experiment(s) the new information relates to. Go to the table/figure that contains that information.
2. Observe how the values associated with the new information compare to values of other trials/samples already described in the appropriate table or figure. See where the new information "fits" among the initial data.
3. Make conclusions about what the new information means.

Why Is This So Challenging?

For most students, the methodical process outlined above is not automatic. One reason is that it's unlike anything they use in school on a daily basis: for the most part, students can safely assume that the graphs they encounter in chemistry or math class will be more straightforward and less demanding. Another reason is that it simultaneously incorporates multiple elements of analysis: novelty, extrapolation, and integration of textual and graphical information. All this under strict timing conditions and the always-present mental pressure of a college admissions exam. Let's also not forget that the ACT Science is the fourth of four sections comprising the test; by the time they hit the Science, students will already have worked through nearly 2.5 hours of testing across the English, Math, and Reading sections. Fatigue is real, and pulling off this kind of reasoning while mentally exhausted is no small feat.

The Upshot

The example above is just one of many I could have highlighted. What's clear is that teaching students about models can help them take questions that seem overwhelming and break them down into manageable chunks. Once students see how a set approach can make the flow of questions much more predictable, they are more likely to get to the point more efficiently, extract the relevant data more quickly, and connect information from multiple sources more successfully. The results are higher scores and happier students.