When several SHSAT ELA choices “all sound right,” the solution is usually not stronger intuition. It is a clearer task and evidence routine.
Use current official practice through the NYC SHSAT Portal and confirm current test information on the NYCPS specialized-high-schools page. The routine below is original study guidance, not an official scoring formula.
Step 1: name the task
Before opening the choices, state what the question asks:
- Main idea or purpose
- Inference
- Detail or evidence
- Tone or attitude
- Meaning in context
- Author's craft or structure
- Revising or editing decision
Different tasks require different evidence. A true detail may not answer a main-purpose question. A plausible inference may be too strong for the text.
Step 2: make a short prediction
Use your own words. The prediction does not need to match the credited answer exactly; it creates a standard for evaluating choices.
For a main-purpose question, complete:
The passage mainly shows/explains/argues ___ by ___ .
For an inference question:
The text most strongly suggests ___ because ___ .
For a revising/editing question:
The sentence needs ___ because ___ .
If you cannot complete the sentence, return to the relevant text before comparing choices.
Keep the prediction proportional to the question. A main-idea prediction may need a full clause; a vocabulary-in-context prediction may be one plain synonym; a transition question may need only “contrast” or “result.” Long predictions can become another reading burden. Their job is to create a target, not to rewrite the passage.
Step 3: point to the evidence
Evidence can be a sentence, a repeated pattern, a contrast, a change, or a structural relationship. Mark it.
Do not say only, “This answer feels supported.” Finish the thought:
This answer is supported by ___, which shows ___ .
For tone, identify both the subject and the language that creates the attitude. For main idea, test whether the evidence represents the whole passage rather than one paragraph.
Step 4: reject choices by flaw type
Tempting wrong answers often fail in repeatable ways:
| Flaw | What it does |
|---|---|
| True but irrelevant | States a real detail that does not answer the task |
| Too broad | Claims more than the passage covers |
| Too narrow | Describes only one example or moment |
| Too strong | Changes may or suggests into proves or always |
| Reversed | Flips cause, attitude, comparison, or sequence |
| Unsupported addition | Introduces a motive or fact not established by the text |
Name the flaw. This is more transferable than memorizing that “choice C was wrong.”
Step 5: use a tie-break sentence
When two choices remain, compare them directly:
Choice A requires me to assume ___, while choice B is supported by ___ .
or
Choice A answers a detail question, while the prompt asks for the paragraph's purpose.
The goal is not to prove that the credited answer is perfect in isolation. It is to show why it answers the exact question better than the alternatives.
Practice with shorter passages first
If the routine breaks down on a long passage, reduce the load. Use one paragraph and one question type. Explain the task, prediction, evidence, and choice flaws aloud or in writing.
Then move to a short passage, a mixed set, and finally a timed section. This builds the reasoning before adding endurance.
Taking more full tests will not automatically repair a blurry concept. If inference remains unclear, isolate inference. If sentence boundaries are unstable, repair the grammar structure before returning to mixed revising/editing sets.
Separate reading errors from pacing errors
A student may use the routine accurately without time pressure but abandon it late in a session. That is different from not understanding the question type.
Track whether the error appears early or only under time. If the reasoning is accurate but slow, practice shorter predictions and more efficient evidence marking. If the reasoning is inaccurate at any speed, reteach the concept.
A compact practice card
Write this on scratch paper during practice until it becomes familiar:
- Task?
- Predict?
- Evidence?
- Why not the others?
- Does my choice answer the exact question?
The routine is not meant to create five long essays per question. With practice, each step becomes shorter.
Measure reasoning, not only accuracy
During review, score the routine itself. Did the student name the task, identify evidence, and reject the tempting choice for a valid reason? A lucky correct answer with no support still needs review. A wrong answer with a sound process but one misread word needs a different fix from a concept gap.
Keep one example of each repeated flaw type in the error log. Before the next session, retrieve the flaw and prevention rule from memory. Then test it on fresh text rather than immediately repeating the same answer choices.
On the adaptive SHSAT, treat each new question as its own decision. Do not let uncertainty from the previous item determine the next one.
