3.2 Prereferral Interventions, Multi-Tiered System of Supports (MTSS) & Response to Intervention (RtI)

3.2 Prereferral Interventions, Multi-Tiered System of Supports (MTSS) & Response to Intervention (RtI)

Quick Summary: Prereferral intervention frameworks are designed to deliver immediate, targeted instructional and behavioral support to struggling learners within the general education environment before any formal referral to the Committee on Special Education (CSE). Under New York State regulations (8 NYCRR Part 100.2[ii] and 200.4), Response to Intervention (RtI) is legally mandated as part of the evaluation process for determining whether a student in grades K–4 has a Specific Learning Disability (SLD) in reading, replacing the historically flawed IQ-achievement discrepancy model with empirical data documenting a dual discrepancy in performance level and learning rate.

Multi-Tiered System of Supports (MTSS) is an overarching prevention and intervention framework that integrates academic supports (Response to Intervention [RtI]) and behavioral/social-emotional supports (Positive Behavioral Interventions and Supports [PBIS]). The central philosophy of MTSS is that all students must receive high-quality, scientifically research-based core instruction, with increasingly intensive layers of targeted intervention matched systematically to demonstrated student need.

---

The MTSS and RtI Architecture

The MTSS framework operates across three distinct, fluid tiers of instructional intensity. Movement between tiers is governed strictly by objective, data-based decision rules derived from universal screening and continuous progress monitoring.

                                  The MTSS / RtI Triangle

                                         /\  Tier 3: Intensive (~1-5%)
                                        /  \  - 1:1 or 1-3 micro-group; daily 45-60 min
                                       /    \ - Weekly CBM progress monitoring
                                      /------\ Tier 2: Targeted Supplemental (~10-15%)
                                     /        \ - Small group (3-5); 3-4x/week, 20-30 min
                                    /          \ - Bi-weekly CBM progress monitoring
                                   /------------\ Tier 1: Universal Core (~80-85%)
                                  /              \ - High-quality core general ed instruction
                                 /                \ - Universal screening 3x/year (Fall, Winter, Spring)
                                /------------------\

Universal Screening Protocols

Universal screening serves as the diagnostic radar of the MTSS model. Administered three times annually (Fall benchmark in September, Winter benchmark in January, and Spring benchmark in May), universal screeners are brief, standardized, reliable instruments administered to all students. Common validated universal screeners include:

• DIBELS 8th Edition (Dynamic Indicators of Basic Early Literacy Skills): Measures phonemic awareness, letter naming, nonsense word fluency, word reading fluency, and oral reading fluency.
• AIMSwebPlus: Assesses early literacy, reading comprehension, vocabulary, early numeracy, and mathematics computation.
• FastBridge / easyCBM: Provides computer-adaptive and teacher-administered probes in reading and mathematics.

Students scoring below the 20th or 25th percentile (or below validated cut-scores) are immediately flagged for Tier 2 intervention review, preventing the historical "wait-to-fail" delay.

---

Curriculum-Based Measurement (CBM) Methodologies

Curriculum-Based Measurement (CBM) is the gold standard for progress monitoring within MTSS and special education. Developed by Stanley Deno and colleagues at the University of Minnesota, CBM differs fundamentally from informal classroom assessments because it utilizes General Outcome Indicators (GOIs)—standardized, brief capstone tasks that reflect broader, cumulative academic competence.

Key CBM Reading Probes

1. Oral Reading Fluency (ORF / Passage Reading Fluency [PRF]):
   • Administration: The student reads aloud from a standardized, unpracticed grade-level passage for exactly 1 minute.
   • Scoring: Evaluator tallies total words read, errors (mispronunciations, substitutions, omissions, hesitations exceeding 3 seconds), and self-corrections (made within 3 seconds, scored as correct).
   • Metric: Words Correct Per Minute (WCPM) and accuracy percentage ($\frac{\text{Words Correct}}{\text{Total Words Read}} \times 100$).
2. Maze Comprehension Probes:
   • Administration: The student silently reads a standardized passage for 2.5 to 3 minutes. The first sentence is intact; beginning in the second sentence, every seventh word is deleted and replaced with a three-word choice: (1) the correct target word, (2) an incorrect word of the same part of speech, and (3) a semantically nonsensical distracter.
   • Scoring: Number of correct selections minus incorrect selections.
   • Metric: Standardized silent reading comprehension index.
3. Early Literacy Probes:
   • Phonemic Segmentation Fluency (PSF): Examiner says a word (e.g., "flag"); student produces individual phonemes (/f/ /l/ /æ/ /g/). Measured in correct sound segments per minute.
   • Nonsense Word Fluency (NWF): Student reads phonetically regular pseudowords (e.g., "mip", "lut"). Measures alphabetic principle and phonics decoding without reliance on sight-word memorization. Scored as Correct Letter Sounds (CLS) and Whole Words Read (WWR).

Key CBM Mathematics Probes

1. Mathematics Computation (M-COMP): Brief, timed probes (typically 2–4 minutes) assessing computational fluency across grade-appropriate operations (addition, subtraction, multiplication, division, fractions, decimals).
2. Mathematics Concepts and Applications (M-CAP): Standardized timed probes (6–10 minutes) assessing word-problem solving, measurement, geometric relationships, time, money, and data analysis.

---

Graphing CBM Data: Baselines, Aimlines, and Trend Lines

CBM data has little instructional utility in raw form; it must be visually graphed to enable empirical decision-making.

                             CBM Progress Monitoring Graph
  WCPM (Words Correct/Min)
    |
 90 |
 80 |                                                      * (Wk 10: 74)
 70 |                                            *---------O Target Goal (70)
 60 |                              * (Wk 6)   .-'   (Aimline)
 50 |                  * (Wk 4)  .-'       .-'
 40 |        * (Wk 2)         .-'       .-'
 30 |   *.-'--------------.-'--------.-'  (Trend line below aimline)
 20 |   X (Baseline Median: 24)
 10 |
  0 +---|---------|---------|---------|---------|---------|---------|----->
     Baseline   Wk 2      Wk 4      Wk 6      Wk 8      Wk 10     Wk 12 (Weeks)

  Key: X = Baseline Median; O = Target Benchmark Goal; Line with O = Aimline; * = Weekly Data

Step-by-Step Graphing Protocol

1. Establish the Baseline:
   • Administer three alternate equivalent forms of the CBM probe on three separate days (or within a single testing session using three different passages).
   • Identify the median score (the middle score) among the three probes, not the mean. The median controls for transient outliers (e.g., scores of 22, 38, and 24 yield a baseline median of 24).
   • Plot the median baseline at "Week 0" on the vertical axis.
2. Establish the Target Goal and Aimline (Goal Line):
   • Determine the end-of-intervention benchmark goal (e.g., achieving 70 WCPM by Week 12).
   • Plot the target goal at the end-date (Week 12).
   • Draw a straight line connecting the baseline median to the target goal. This is the Aimline—representing the required weekly Rate of Improvement (ROI) or expected slope of learning: $\text{Required ROI (Slope)} = \frac{\text{Target Goal} - \text{Baseline Median}}{\text{Total Weeks of Intervention}}$
3. Plot Weekly Progress Monitoring Data:
   • Administer an alternate equivalent CBM form every 1 to 2 weeks under identical standardized conditions and plot each score.
4. Calculate the Empirical Trend Line:
   • Using the Split-Middle Method (Tukey method) or linear regression, calculate the student's actual growth slope across data points to compare the empirical trend line directly against the aimline.

---

Data-Based Decision Rules in MTSS/RtI

Intervention teams must adhere to predefined, objective decision rules to determine whether to maintain, modify, intensify, or fade an intervention.

The 4-Point Rule (Decision Rules)

• Rule 1: Insufficient Progress (Change Intervention): If four consecutive data points fall below the aimline, the student is not making adequate progress to achieve the goal. The intervention team must immediately convene to verify treatment fidelity and modify the intervention.
• Rule 2: Accelerating Progress (Raise Goal or Fade): If four consecutive data points fall above the aimline, the student's rate of growth significantly exceeds expectations. The team should either increase the end-of-year goal (raise the aimline) or formulate a systematic plan to step down intervention intensity toward Tier 1 core instruction.
• Rule 3: Inconsistent / Variable Data (Examine Extraneous Factors): If data points oscillate erratically above and below the aimline without a clear trajectory, the team maintains the intervention for an additional 2 to 3 weeks while auditing environmental disruptions, student attendance, and tester consistency.

Treatment Fidelity (Implementation Integrity)

Core Principle: An intervention cannot be judged ineffective if it was not delivered as designed.

Before modifying an intervention or referring a student to the CSE, the intervention team must audit treatment fidelity across four core dimensions:

1. Adherence: Were all evidence-based intervention components delivered precisely according to the curriculum manual and instructional protocol?
2. Dosage / Exposure: Did the student receive the prescribed number of sessions, minutes per session, and total weeks (accounting for absences, holidays, or assemblies)?
3. Quality of Delivery: Was the intervention delivered by a certified, trained professional using explicit, systematic pedagogical modeling and corrective feedback?
4. Student Responsiveness: Was the student actively engaged during the intervention sessions?

Treatment fidelity is verified through direct observation checklists, intervention logs, and student attendance tracking.

---

RtI in Specific Learning Disability (SLD) Identification: The Dual Discrepancy Model

Shortcomings of the Historical IQ-Achievement Discrepancy Model

Prior to the 2004 reauthorization of IDEA, SLD identification relied almost exclusively on the severe discrepancy model, which required a statistically significant difference (typically 1.5 to 2.0 standard deviations, or 22 to 30 points) between a student's Full Scale IQ and standardized achievement test scores. This model suffered from fatal clinical flaws:

• The "Wait-to-Fail" Flaw: Severe discrepancies rarely manifest until third or fourth grade, after years of academic failure, eliminating early intervention during critical developmental windows.
• Measurement Artifacts: IQ scores are not independent of reading ability; reading deficits suppress verbal IQ, artificially closing the discrepancy over time.
• Cultural and Linguistic Bias: Discrepancy formulas disproportionately misidentified culturally and linguistically diverse learners or failed to identify students with high intelligence whose achievement was suppressed to the average range.
• No Instructional Utility: The discrepancy formula provides zero information regarding what specific instructional strategies or interventions the student requires.

New York State Regulations: 8 NYCRR Part 100.2(ii) and Part 200.4

In New York State, regulations governing RtI are explicitly codified:

• Mandatory RtI in Reading (Grades K–4): Under 8 NYCRR 100.2(ii), all public school districts in New York State must establish an approved RtI framework for students in grades kindergarten through fourth grade in the area of reading.
• Mandatory Submission of RtI Data: When a student in grades K–4 is referred to the CSE for suspected Specific Learning Disability in reading, the district must include comprehensive RtI data demonstrating that the student received high-quality instruction and targeted interventions in the general education setting.

The Dual Discrepancy Framework (Research Model, Not a New York Regulation)

Be precise about where this idea comes from, because the exam rewards candidates who can separate regulation from research. New York's regulations never use the phrase "dual discrepancy." What 8 NYCRR § 200.4(j)(3)(i) does is give the CSE two permitted routes to document a learning disability: (a) the student "does not make sufficient progress to meet age or State-approved grade-level standards ... when using a process based on the student's response to scientific, research-based intervention pursuant to section 100.2(ii)," or (b) the student exhibits a pattern of strengths and weaknesses relevant to identifying a learning disability. 8 NYCRR § 200.4(j)(4) then adds the hard prohibition: effective on and after July 1, 2012, a district shall not use severe discrepancy criteria to determine that a student in grades K–4 has a learning disability in reading.

Dual discrepancy is the research operationalization of route (a), developed by Fuchs and Fuchs and widely used in RtI implementation. It is how practitioners show "insufficient response" with data, not a separate legal test:

                                The Dual Discrepancy Model

        Discrepancy 1: Academic Level              Discrepancy 2: Rate of Learning (Slope)
    +------------------------------------+    +--------------------------------------------+
    | Performance Level is substantially |    | Rate of Improvement (ROI) is substantially |
    | below normative peer benchmarks    |    | flatter than peers despite intensive,      |
    | (e.g., < 10th-16th percentile).    |    | high-fidelity Tier 2 & Tier 3 instruction. |
    +------------------------------------+    +--------------------------------------------+
                                       \        /
                                        \      /
                                         \    /
                              +--------------------------+
                              |    SPECIFIC LEARNING     |
                              |   DISABILITY ELIGIBILITY |  (Prerequisite: Ruling out
                              |      DOCUMENTATION       |   exclusionary factors)
                              +--------------------------+

1. Discrepancy in Performance Level: The student's current academic performance level is severely discrepant from normative grade-level expectations (typically falling below the 10th or 16th percentile on validated CBM or standardized achievement measures).
2. Discrepancy in Rate of Learning (Growth Slope): The student's rate of improvement (ROI) over time, represented by their CBM trend line slope, is significantly lower than the growth rate exhibited by typical peers and inadequate to close the achievement gap, despite sustained, high-fidelity Tier 2 and Tier 3 interventions.

---

NY Classroom Case Scenario: CBM Data Analysis & Team Referral Decision

Case Profile: Sofia (Grade 2, Albany Public Schools)

Sofia was flagged during the Fall universal screening in Grade 2 with an Oral Reading Fluency (ORF) score of 14 WCPM (benchmark was 52 WCPM, placing her at the 4th percentile). The school's MTSS team placed Sofia in a 12-week Tier 2 supplemental reading intervention: Wilson Fundations small-group instruction (4 students, 30 minutes daily, 4 days per week) targeting phonological awareness and syllable decoding.

Sofia's 12-Week CBM Progress Monitoring Data Log

• Baseline Median: Administered three probes: 12, 16, 15 WCPM $\rightarrow$ Baseline Median = 15 WCPM.
• Target Benchmark Goal: 45 WCPM by Week 12.
• Required Aimline Slope: $\frac{45 - 15}{12} = +2.5\text{ WCPM/week}$.

MTSS Team Referral Determination

At the end of Week 12, the MTSS team evaluated Sofia's cumulative data:

1. Performance Level Discrepancy: Sofia achieved 24 WCPM at the end of Week 12, compared to the grade-level winter benchmark of 72 WCPM (she remains below the 2nd percentile).
2. Rate of Learning Discrepancy: Over 12 weeks of intensive, high-fidelity intervention, Sofia's empirical growth rate was only +0.75 WCPM per week ($\frac{24 - 15}{12}$), compared to the expected peer growth rate of +1.8 to +2.0 WCPM per week and her target aimline slope of +2.5 WCPM per week.
3. Fidelity Verification: Intervention logs confirmed 96% session attendance and 100% adherence to instructional protocols.
4. Action: The team initiates a formal written referral to the Committee on Special Education (CSE) for suspected Specific Learning Disability (Dyslexia), compiling the 12-week CBM graphs, fidelity checklists, and intervention artifacts as legally mandated under 8 NYCRR Part 100.2(ii).