How to Write Reading Comprehension IEP Goals

Quick Answer

A strong reading comprehension goal defines the passage level, reading condition, response type, and criterion. Measure literal/inferential question accuracy or a stable retell rubric across comparable passages. Keep decoding or writing supports explicit so changes in those demands do not get mistaken for comprehension growth.

In reading comprehension data collection, a strong draft begins on the data sheet, not in the goal bank. For reading comprehension, the core measurement is percent correct on defined literal/inferential questions or a stable retell rubric.

What a strong reading comprehension goal includes

For comprehension, the condition determines what the score means. In Condition → Behavior → Criterion, name the passage level or type, whether the student reads independently or hears the text, and the response demand. The behavior can be answering literal or inferential questions, citing evidence, identifying main idea, or producing a retell. The criterion should use a consistent question set or anchored rubric across repeated trials rather than an undefined classroom percentage.

Reading-comprehension goals fall within the measurable annual goals required by 34 CFR §300.320(a)(2). There is no federal comprehension cutoff. Keep text level, question type, response mode, and allowed supports stable enough that the criterion can be interpreted across data points.

Question set — % correct by item type: Best when literal/inferential demands can be balanced. Retell — points on anchored rubric: Useful when oral summary is the target. Written response — rubric score + required evidence: Useful for older students when reading and writing demands are intentionally controlled.

Define the comprehension probe beyond a percentage: passage level and genre, whether the student reads independently or hears the text, the number and type of questions, and the response format. An '80% comprehension' score is not comparable if one probe contains five literal questions and the next requires written inference and evidence.

Comprehension asks what meaning the student constructs from text. Fluency data can help explain performance, but WCPM is not a comprehension score.

To keep comprehension separate from rate and connect the target to present levels, use reading fluency goals, write a PLAAFP that supports the goals, Condition–Behavior–Criterion formula, and browse a full reading goal bank.

Build comparable question sets instead of chasing a single percentage

A comprehension percentage is only as stable as its question blueprint. If one probe contains mostly literal recall and another contains inference, vocabulary-in-context, and main-idea questions, an 80% score on each does not represent the same demand. Create a small blueprint—such as equal literal and inferential items—and reuse it with parallel passages.

For retell goals, anchor each rubric point to observable features such as naming the central idea, including key events or details, preserving sequence, and stating a reasonable inference. That keeps scoring from drifting toward a teacher's general impression of whether the retell “sounded good.”

How to get a baseline

Build the baseline from comparable observations instead of one convenient classroom moment. For reading comprehension, useful sources include:

Two or three leveled passages with matched question sets. A retell scored with the same story-grammar or informational-text rubric. Error analysis separating literal, inferential, vocabulary, and main-idea items.

Use leveled passages with matched question sets and keep the response mode stable. A second baseline source can be a retell scored with the same story-grammar or informational-text rubric. Error analysis should separate literal recall, inference, vocabulary-in-context, main idea, and evidence use; those patterns help the team choose one teachable comprehension target instead of a catch-all goal.

For each comprehension sample, record text level, genre, access mode (independent reading, audio, or another documented support), question mix, and scoring rule. Separate literal, inferential, main-idea, or retell data when the goal targets one of those constructs specifically.

Name the text level, whether reading is independent or supported, and the question mix. '80% comprehension' is meaningless if one probe is five literal questions and the next is a written inference task.

Example reading comprehension goals

These comprehension goals illustrate different constructs—literal recall, inference, main idea, evidence use, and retell. Select only the construct supported by the student's baseline and keep passage level, response mode, supports, and scoring stable enough for later comparisons. Individualize the comprehension goal from the student's own passage-response data.

K–2 — Given a 150–250 word instructional-level passage and 5 teacher-created questions, the student will answer literal comprehension questions with at least 80% accuracy across 3 consecutive probes. (Condition: a 150–250 word passage and 5 questions | Behavior: answer literal comprehension questions | Criterion: at least 80% across 3 consecutive probes) How to measure: score parallel question sets; keep question type constant.

K–2 — After listening to or reading a short narrative, the student will retell the setting, problem, two key events, and resolution with 4 of 5 rubric points on 3 samples. (Condition: a short narrative | Behavior: retell required story elements | Criterion: 4 of 5 rubric points on 3 samples) How to measure: use the same retell rubric for each sample.

3–5 — Given a grade-level passage and 8 mixed literal and inferential questions, the student will answer at least 6 correctly on 3 of 4 probes. (Condition: a grade-level passage and 8 questions | Behavior: answer literal and inferential questions | Criterion: 6 of 8 on 3 of 4 probes) How to measure: parallel passages with balanced question types.

3–5 — Given an informational passage, the student will identify the main idea and two supporting details with at least 80% rubric accuracy across 4 samples. (Condition: an informational passage | Behavior: identify main idea and two supporting details | Criterion: at least 80% across 4 samples) How to measure: score with a 5-point response rubric.

6–8 — Given a grade-level text, the student will cite one relevant piece of textual evidence for an inference in 4 of 5 responses across 3 probes. (Condition: a grade-level text | Behavior: support an inference with relevant evidence | Criterion: 4 of 5 responses across 3 probes) How to measure: use short constructed-response probes with a defined rubric.

9–12 — Given a content-area passage, the student will summarize the central idea and two essential details without adding inaccurate information, earning at least 4 of 5 rubric points on 3 consecutive samples. (Condition: a content-area passage | Behavior: write an accurate summary | Criterion: 4 of 5 rubric points on 3 consecutive samples) How to measure: score each summary with the same analytic rubric.

Control the question mix so percentages mean something

“80% comprehension” is only interpretable when the denominator represents a stable kind of task. A five-question probe with four literal questions and one inference is not equivalent to a probe dominated by inference, main idea, and text-evidence questions. Build parallel sets or record subskill scores so the team can see whether growth is broad or limited to one question type.

Also decide how the student accesses the passage. If the goal is comprehension rather than decoding, audio or adult-read text may be an appropriate condition for some students when that matches the IEP. If the student must read independently, state that. Changing access between baseline and progress probes can make comprehension scores move even when comprehension itself has not changed.

How to write the matching present level

The PLAAFP should report current comprehension under the same access condition the goal will use and name the educational effect—for example, accurate recall of explicit details but weak inference that limits independent understanding of grade-level text.

Worked present-level example: Across three sixth-grade passages read independently, Luis answered 6/10, 5/10, and 6/10 questions; inferential items averaged 2/5 while literal items averaged 4/5. He identifies stated details more reliably than implied relationships.

Keep the question mix and passage demand comparable across formal probes. If the team changes from oral responses to written responses or from supported to independent reading, note that new demand instead of treating the scores as one uninterrupted series.

Progress monitoring

Use a simple routine staff can sustain; inconsistent data are less useful than a modest schedule followed well. Use parallel passages and the same question blueprint every 2–3 weeks. Keep literal and inferential subscores if the goal names both so growth in one area does not mask a persistent weakness in the other. Before rewriting the goal, inspect whether decoding load, vocabulary, text complexity, or written-response demands changed. A flat comprehension line can come from inconsistent probes rather than stagnant comprehension.

Use the same comprehension metric through the reporting period—such as percent correct on matched literal/inferential question sets or a stable retell rubric. 34 CFR §300.320(a)(3) requires the IEP to specify measurement and periodic reporting. Classroom probes may be more frequent than family reports, but the report should combine comparable passage/question data rather than substitute a one-day class grade.

Common mistakes

Watch for these drafting traps:

Writing 'will understand grade-level text' with no observable response. Changing question types from probe to probe. Scoring retells without an anchored rubric. Letting decoding difficulty contaminate a comprehension-only goal without stating supports. Combining comprehension and fluency in one mastery number. Using classroom grades as the sole progress measure.

Try the scoring rule on one passage and question set with a colleague. Agreement should be possible on literal versus inferential items, acceptable responses, and rubric points without adding explanations that are absent from the written goal.

FAQ

Should comprehension goals use percent correct or a retell rubric?

Choose the measure that matches the reading behavior you actually need to improve. Percent correct works well for clearly defined literal, inferential, main-idea, or evidence questions. A retell rubric is better when the target is organized recall or synthesis. Whichever tool you choose, use the same scoring structure for baseline and progress monitoring.

Can I combine literal and inferential questions in one comprehension goal?

Yes, if the goal clearly defines the mix and the student needs a broad comprehension outcome. However, a single percentage can hide a large gap between literal and inferential performance. When that distinction matters instructionally, report the categories separately or set a criterion for each question type so the team can see where progress is occurring.

How do I control passage difficulty when measuring comprehension?

Record the text level, length, genre, and support conditions used for each probe. Progress data are hard to interpret if one week uses a short narrative with pictures and the next uses dense informational text. Use comparable passages or a stable rubric, and document any read-aloud, vocabulary preview, or annotation support that is part of the condition.

What belongs in the baseline for a reading comprehension goal?

Include the student’s current score using the same type of passage and question set that will be used to judge the annual goal. A useful baseline might state percent correct across three leveled passages, plus a breakdown showing stronger literal than inferential performance. That makes the instructional need and the future criterion directly comparable.

Should a comprehension goal name a specific reading program?

Usually no. The goal should describe the student behavior and measurable outcome, not lock the IEP to a commercial program unless the team has a specific reason to do so. Name the text condition, question type, response format, and criterion. Instructional materials can change while the underlying skill and progress measure remain stable.

Need goals customized to your student?

Paste your assessment data and generate an IDEA-compliant, individualized goal in seconds.

Try it free →