Command of Evidence: Textual questions ask you to connect a claim with relevant evidence. Depending on the task, a choice may be a quotation, a finding that would support a conclusion, or information that would weaken it. The answer is not always an existing sentence within a few lines of an underlined claim.
Start with the direction of the task
Read whether the question asks for support, an illustration or a challenge. Summarise the claim and identify the relationship it predicts. Then consider what each answer choice would do to that relationship. Topic overlap alone is not evidence.
An original practice example
A researcher proposes that shading reduces the rate at which water evaporates from identical shallow containers. Which finding would support that proposal? If shaded and unshaded containers start with equal water, receive no additional water and are otherwise treated alike, consistently smaller water losses in the shaded group would support the proposal. Finding that shaded containers are painted a different colour would not by itself test evaporation; a larger water loss in the shaded group would work against the proposal under those conditions.
The example depends on comparing like conditions and identifying the measured outcome. It does not require a supporting sentence hidden elsewhere in a passage. For a literary quotation task, the same discipline applies: connect the quoted behaviour or language with the exact claim rather than a nearby theme.
A four-step routine
- Identify the claim and the requested direction: support or weaken.
- Predict the kind of evidence that would affect that claim.
- Evaluate each option’s logical relationship to the claim.
- Reject options that are merely relevant to the topic, introduce a different population or reverse the comparison.
Practise without invented score guarantees
Log why each wrong option fails, then try a new question that uses the same reasoning pattern. A fixed three-line search rule, guaranteed two-week accuracy gain or exact count of Command of Evidence questions is not a reliable basis for a study plan.
Reading and Writing uses short texts or text pairs followed by one question. Similar skills are grouped and questions are arranged from easier to harder within those groups. Strong first-module performance leads to a more challenging second-module mix; it does not earn easier second-module questions. Both modules contribute to the section score.
Use the official Reading and Writing overview, Student Question Bank and scoring explanation for current examples and format details.
Frequently asked
- Are answers always sentences already present in the passage?
- No. A task may ask which hypothetical finding would support or weaken a claim, or which quotation best supports it.
- Does strong Module 1 performance lead to easier Module 2 questions?
- No. It leads to a more challenging mix in Module 2; both modules contribute to the section score.
- Is there a guaranteed two-week accuracy gain?
- No. Track performance on new questions and use your own error pattern to plan practice.
Sources
Start with a diagnostic to find out which of these applies to you, or see how tutoring works.