5.3 Literacy Assessment Systems: Assessment Types, IRIs, Error Analysis & Communicating Results

Literacy Assessment Systems: Diagnostic, Formative, Summative & IRI

Assessment is the engine that drives effective literacy instruction. Without valid, reliable, and continuous assessment data, teaching becomes guesswork. A comprehensive literacy assessment system integrates multiple assessment types—screening, diagnostic, formative, and summative measures—to provide a holistic profile of each student's reading and writing development. Elementary educators must master assessment typologies, interpret qualitative and quantitative metrics, and administer diagnostic tools like Informal Reading Inventories (IRIs) and running records to make precise instructional decisions.

---

Comprehensive Literacy Assessment Typologies

Assessments are classified by their design, reference framework, and administrative purpose. Understanding these classifications ensures that teachers select the right assessment tool for the right instructional question.

Norm-Referenced vs. Criterion-Referenced Assessments

Assessment Functions in a Multi-Tiered System

1. Universal Screening: Brief, standardized assessments administered to 100% of students three times per year (fall, winter, spring). Screeners act as an early warning system to identify students at risk of reading difficulties before academic failure occurs (e.g., DIBELS, AIMSweb, FastBridge). Measures focus on critical foundational indicators such as phonemic segmentation fluency, letter naming fluency, nonsense word decoding, and oral reading fluency.
2. Diagnostic Assessments: In-depth, targeted evaluation tools administered individually to students flagged by universal screeners or teacher observation as struggling. While a screener indicates that a student is behind, a diagnostic assessment reveals why. Diagnostic tools deconstruct reading into component subskills—phonological awareness, phonics patterns, multisyllabic decoding, sight word vocabulary, oral vocabulary, and listening comprehension—to pinpoint exact instructional entry points (e.g., Comprehensive Test of Phonological Processing [CTOPP], Diagnostic Reading Inventory).
3. Formative Assessments: Continuous, low-stakes assessments for learning conducted during the instructional process. Formative assessments provide immediate feedback to both teachers and students, enabling real-time adjustments to lesson pacing, scaffolding, and grouping. Examples include teacher observational notes, oral retellings, running records, white-board checks, conferring notes, and daily exit tickets.
4. Summative Assessments: High-stakes assessments of learning administered at the conclusion of an instructional unit, grading period, or school year. Summative assessments measure cumulative learning, determine final grades, and evaluate curriculum effectiveness. Examples include end-of-unit reading comprehension exams, published research projects evaluated with rubrics, and state standardized exams.

---

The Informal Reading Inventory (IRI) & Reading Levels

An Informal Reading Inventory (IRI) is an individually administered diagnostic assessment designed to evaluate multiple facets of a student's reading performance. A standard IRI consists of two core components:

1. Graded Word Lists: Lists of 10 to 20 words per grade level read in isolation to evaluate rapid sight word recognition and decoding automaticity, establishing an entry point for passage reading.
2. Graded Reading Passages: Leveled narrative and informational passages that the student reads orally and silently while the teacher records oral reading accuracy, reading rate (words correct per minute), and answers to comprehension questions.

Mathematical Calculations for IRI

To determine a student's functional reading level, the teacher calculates two percentages from the oral passage reading:

$\text{Word Recognition Accuracy \%} = \left( \frac{\text{Total Words in Passage} - \text{Uncorrected Miscues}}{\text{Total Words in Passage}} \right) \times 100$

$\text{Comprehension \%} = \left( \frac{\text{Number of Correct Comprehension Answers}}{\text{Total Comprehension Questions Asked}} \right) \times 100$

Critical IRI Rule: Self-corrections (SC)—errors that the student corrects independently within three seconds—are recorded for qualitative analysis but are never counted as quantitative errors in calculating accuracy! Hesitations, repetitions, and self-corrections reflect active self-monitoring, not decoding failure.

The Three Functional Reading Levels

Based on these calculations, the student's performance on that specific text level is classified into one of three reading levels:

1. Independent Reading Level:
   • Word Recognition Accuracy: 95% to 100%
   • Comprehension: 90% to 100%
   • Instructional Application: The text is effortless for the student. The reader decodes automatically, reads with expressive prosody, and comprehends fully without adult assistance. Ideal for sustained silent reading (SSR), recreational library reading, independent research, and building reading stamina.
2. Instructional Reading Level:
   • Word Recognition Accuracy: 90% to 94%
   • Comprehension: 75% to 89%
   • Instructional Application: This is the student's Zone of Proximal Development (ZPD). The text provides sufficient challenge to require cognitive effort, yet enough accessibility that the student can succeed with teacher guidance, vocabulary pre-teaching, and strategy scaffolding. This is the optimal level for guided reading groups and teacher-directed close reading instruction.
3. Frustration Reading Level:
   • Word Recognition Accuracy: Below 90% OR Comprehension: Below 75%
   • Instructional Application: The text is too difficult. Frequent decoding breakdowns overload working memory, resulting in halting, word-by-word reading, high cognitive fatigue, anxiety, and a collapse in comprehension. Texts at this level must never be assigned for independent or small-group guided reading (though they may be used as teacher read-alouds to develop listening comprehension and academic vocabulary).

---

Comparison Table: Reading Levels IRI Criteria

---

Running Records & Oral Reading Error Analysis

A running record is an ongoing formative assessment protocol developed by Marie Clay to document a student's oral reading behaviors on authentic text. The teacher sits beside the student with a blank sheet or text copy, using standardized shorthand symbols to record correct words (checkmarks), substitutions, omissions, insertions, repetitions, and self-corrections.

Coding Errors: The MSV Labels You Will Still See

Many running-record forms ask teachers to code each substitution as meaning (M), structure (S), or visual (V). You should recognize these labels, because they describe what kind of error a student made. In Florida, however, they may not become the basis for teaching word reading: s. 1001.215(7) and s. 1008.25, F.S., make phonics for decoding and encoding the primary strategy for teaching word reading and bar the three-cueing model. Use the labels for diagnosis only:

                    [ Meaning / Semantic (M) ]
                     "Does it make sense?"
                              / \
                             /   \
                            /     \
                           /       \
                          /         \
 [ Structure / Syntactic (S) ] <---> [ Visual / Graphophonic (V) ]
     "Does it sound right?"            "Does it look right?"

1. Meaning / Semantic Cues (M):
   • Guiding Question: Does the error make sense in the context of the sentence and story?
   • Linguistic Source: Background knowledge, context clues, pictorial illustrations, narrative comprehension.
   • Example: The text reads: "The horse ran across the green pasture." The student reads: "The horse ran across the green field." The word field makes complete sense in this context. The student is reading for meaning (M), but neglected visual cues because field does not look like pasture.
2. Structure / Syntactic Cues (S):
   • Guiding Question: Does the error sound right grammatically within English sentence structure?
   • Linguistic Source: Intuitive grammar, syntax, parts of speech, word order rules.
   • Example: The text reads: "The majestic eagle flew high above the mountain." The student reads: "The majestic eagle soared high above the mountain." Soared is a past-tense verb that seamlessly fits the grammatical slot of flew. The substitution sounds right grammatically (S) and makes sense (M), but does not match visually.
3. Visual / Graphophonic Cues (V):
   • Guiding Question: Does the error look like the printed word? Does it match letter-sound (phoneme-grapheme) correspondences?
   • Linguistic Source: Phonics knowledge, visual word recognition, initial/medial/final letter patterns, word length.
   • Example: The text reads: "The chef poured the milk into the bowl." The student reads: "The chef poured the mild into the bowl." The word mild shares the first three letters (m-i-l) with milk and has a similar shape, indicating strong visual cue reliance (V). However, mild makes no semantic sense and violates English syntax in this position.

Interpreting Error Patterns Under Florida's Decoding-First Rules

Recording what a student said versus what was printed remains valuable diagnostic information. What Florida law changes is the instructional response, which must be phonics-based:

• Meaningful substitutions that do not match the print (pony for stallion, field for pasture): the student is predicting from context instead of decoding. Next step: explicit phonics and word study on the patterns missed (blends, vowel teams, multisyllabic chunks), with prompts such as "Look at all the letters. Say the sounds. Now blend them."
• Look-alike substitutions that lose meaning (captain for carpet): the student uses some letters but does not decode all the way through or notice that meaning broke down. Next step: teach systematic left-to-right decoding and syllable chunking, then have the student reread the sentence to confirm that the decoded word makes sense.
• Self-corrections: evidence of healthy self-monitoring. Praise the student for going back to the letters.

Avoid prompts such as "Look at the picture" or "What word would make sense here?" as word-reading strategies. Those are three-cueing moves, and on the FTCE they mark a wrong answer.

---

Communicating Assessment Results to Students and Families

Competency 4 also expects teachers to interpret formal and informal results for students and stakeholders, not just for their own planning.

• With students: Hold brief data conferences. Show the student a progress-monitoring graph or rubric, name one strength, and set one specific, reachable goal (for example, "Read the vowel-team words in your book without skipping them"). Students who understand their goals improve faster than students who only receive a grade.
• With families: Translate scores into plain language. A percentile rank of 40 means the child scored as well as or better than 40% of the norm group, which is not the same as 40% correct. A grade-equivalent score of 5.2 earned by a third grader means the child did as well on third-grade material as a typical fifth grader in the second month would; it does not mean the child is ready for fifth-grade books. Always pair scores with work samples and the next instructional steps.
• Florida's notification rule: Under s. 1008.25(5), F.S., the parent of a K–3 student identified with a substantial deficiency in reading must be notified in writing immediately. The notice must explain the difficulty in understandable terms, describe current services and the planned intensive interventions, explain the grade 3 promotion requirement (a Level 2 or higher on the statewide ELA assessment unless a good-cause exemption applies), and provide a read-at-home plan with strategies the family can use.
• With colleagues: Bring screening and progress-monitoring data to grade-level and MTSS problem-solving meetings. Keep results confidential, and share them only with people who have a legitimate educational interest, as required by FERPA.

---

Rubrics and Portfolio Assessments

Standardized multiple-choice tests cannot measure expressive literacy, synthesis, or creative craft. Elementary teachers employ authentic, performance-based assessment tools:

Analytic vs. Holistic Rubrics

• Analytic Rubrics: Deconstruct a complex task into separate, distinct criteria (e.g., Content/Ideas, Organization, Voice, Word Choice, Sentence Fluency, Conventions) and score each criterion individually on a qualitative continuum (e.g., Exemplary, Proficient, Developing, Beginning). Analytic rubrics provide granular, actionable diagnostic feedback, showing students exactly where to focus revision.
• Holistic Rubrics: Provide a single comprehensive score (e.g., 1 to 4) based on an overall impression of the work as a unified whole. Holistic rubrics are faster to score for large-scale summative grading, but offer limited diagnostic value for guiding individual student improvement.

Portfolio Assessments

A portfolio assessment is a purposeful, systematic collection of student work gathered longitudinally over time. Portfolios typically contain diverse artifacts: early drafts alongside polished final pieces, audio recordings of oral reading fluency, reading logs, self-reflection sheets, and personal goal-setting plans. Portfolios celebrate authentic growth, document progress across genres, and develop metacognitive self-evaluation as students reflect on their development as readers and writers.

---

Worked Analysis Example: Calculating IRI Levels and Analyzing Miscues

Classroom Context and IRI Protocol

A third-grade teacher administers a 150-word oral reading passage to a student named David. The teacher uses a running record protocol and follows up with 5 comprehension questions.

Observational Data from the Running Record

• Total words in passage: 150 words
• Uncorrected miscues: 9 errors
  • Line 2: Text = "stumbled" -> David read = "stepped" (M: Yes, S: Yes, V: Partial)
  • Line 5: Text = "whispered" -> David read = "whistled" (M: No, S: Yes, V: Strong)
  • Line 8: Text = "frightened" -> David read = "scared" (M: Yes, S: Yes, V: None)
  • Line 11: Text = "lantern" -> David read = "ladder" (M: No, S: Yes, V: Strong)
  • Line 14: Text = "creaked" -> David read = "cracked" (M: Yes, S: Yes, V: Strong)
  • Line 18: Text = "shadows" -> David read = "shades" (M: Yes, S: Yes, V: Partial)
  • Line 21: Text = "glanced" -> David omitted word entirely
  • Line 25: Text = "mysterious" -> David read = "misty" (M: Partial, S: No, V: Partial)
  • Line 28: Text = "trembled" -> David read = "tripped" (M: No, S: Yes, V: Partial)
• Self-corrections: 3 self-corrections (David read "house" for "home", "can" for "could", and "black" for "dark", but immediately caught himself and reread them correctly within 2 seconds).
• Comprehension check: David answered 4 out of 5 comprehension questions correctly (80%).

Step-by-Step Quantitative Calculation

1. Determine Quantifiable Miscues: David made 9 uncorrected miscues. The 3 self-corrections are recorded for qualitative analysis but do not count as errors.
2. Calculate Word Recognition Accuracy: $\text{Accuracy} = \left( \frac{150 - 9}{150} \right) \times 100 = \left( \frac{141}{150} \right) \times 100 = 94.0\%$
3. Calculate Comprehension Accuracy: $\text{Comprehension} = \left( \frac{4}{5} \right) \times 100 = 80.0\%$
4. Determine Reading Level:
   • Accuracy of 94.0% falls directly within the 90% to 94% instructional band.
   • Comprehension of 80.0% falls directly within the 75% to 89% instructional band.
   • Conclusion: This passage places David at his Instructional Reading Level.

Qualitative Diagnostic Synthesis

• David demonstrates active self-monitoring, as evidenced by his 3 self-corrections when syntax or meaning broke down.
• On several miscues (scared for frightened, stepped for stumbled), David relies heavily on meaning (M) and syntax (S), substituting synonyms while ignoring visual print details.
• On other miscues (ladder for lantern, tripped for trembled), David uses only the first letters and does not decode through the rest of the word, then fails to notice that the sentence no longer makes sense.
• Instructional Plan: Provide small-group instruction at this level. Focus on multisyllabic decoding (breaking r-controlled and vowel-team words into syllables) and prompt him to decode through the whole word before rereading for meaning ('Look at all the letters: l-a-n-t-e-r-n. Blend the syllables. Now reread the sentence.').

---

Authentic Classroom Scenarios

Scenario 1: Using Universal Screeners to Guide Diagnostic Testing

At the beginning of second grade, a school administers DIBELS Oral Reading Fluency as a universal screener. A student, Maya, scores in the 14th percentile (well below the benchmark of 52 words correct per minute). The universal screener alerts the teacher to risk, but does not identify the underlying cause. Is Maya struggling with phonemic awareness, phonics decoding, sight word automaticity, or language comprehension? The teacher immediately administers a diagnostic phonics inventory (such as the Quick Phonics Screener). The diagnostic tool reveals that Maya decodes short-vowel CVC words fluently, but struggles with consonant digraphs (th, sh, ch) and long-vowel silent-e patterns. The teacher uses this diagnostic data to design a targeted Tier 2 phonics intervention.

Scenario 2: Running Record Self-Monitoring and Self-Correction

A first-grade teacher listens to a student read a leveled book during a running record. The text reads: "The farmer milked the cow in the barn." The student reads: "The farmer milked the crow in the barn." The student pauses, frowns, looks back at the word, blends the sounds /k/ /ow/, and rereads the sentence: "The farmer milked the... cow in the barn!" In the running record, the teacher notes the initial substitution and marks it as a self-correction (SC). During the subsequent conference, the teacher praises the student's metacognitive awareness: "I loved how you noticed that 'milked the crow' didn't make sense, went back to the letters, and sounded out the word correctly. Checking the letters is what great readers do!"

Scenario 3: Implementing Analytic Rubrics for Writing Self-Assessment

A fourth-grade teacher prepares students to write opinion essays on school lunch options. Rather than grading finished pieces with a single letter grade, the teacher introduces a 4-point analytic rubric with four criteria: (1) Claim and Organization, (2) Evidence and Reasoning, (3) Word Choice and Voice, and (4) Conventions. During mid-unit conferences, students use the rubric to self-evaluate their drafts, highlighting where their evidence meets 'Level 3' and setting a specific goal to reach 'Level 4' by adding peer survey data and transition words. This transparent rubric shifts assessment from a punitive event into an instructional roadmap.