Reading and Language Intervention Barbara Foorman, Ph.D. Florida Center for Reading Research Florida State University What is the Issue? • 33% below basic on G4th NAEP (53% Blacks; 50% Hispanics) ; 17.5% of students are RD • NCLB requires that students at-risk for reading disability receive intervention • The state of the art in reading remediation is prevention and early intervention • IDEA 2004 allows up to 15% of special education funds to be used to provide intervention to struggling readers before they fail to meet grade-level achievement standards. Landmark Studies • Classroom prevention (Foorman et al., 1998, 2006; Connor et al., 2007) • Early intervention (Vellutino et al., 1996; 2003) • Intensive intervention (Torgesen et al., 2001) A G ro w th In W o rd R e a d in g R a w S c o re s B y C u rric u lu m 16 D ire c t C o d e Ins truc tio n E m b e d d e d C o d e Ins truc tio n 14 Im p lic it C o d e - R e s e a rc h Ins truc tio n Im p lic it C o d e - S ta nd a rd Ins truc tio n N u m b e r o f W o rd s 12 10 8 6 4 2 0 O c to be r D e c e m be r F e brua ry April S c h o o l Ye a r B P re d ic te d G ro w th In W o rd R e a d in g S c o re s B y C u rric u lu m 16 D ire c t C o d e Ins truc tio n 14 E m b e d d e d C o d e Ins truc tio n Im p lic it C o d e - R e s e a rc h Ins truc tio n Im p lic it C o d e - S ta nd a rd Ins truc tio n N u m b e r o f W o rd s 12 10 8 6 4 2 0 O c to be r D e c e m be r F e brua ry S c h o o l Ye a r April oo k an Ora d l P L P rin an ho t gu ne A w ag m a e ic re Le A ne tte wa ss A lp r R ren ha e e be co ss t gn S ic I iti tru ns on ct tru ur al ctio A n n W al y or si d s W P re Vo or vi ca k e b S win ula pe g ry lli ng a B R o ea i di R n C ok ng ea o d n R ea Th ing tex di eir B t ng O oo w ks C om n W pr rit eh in en g si o S n pe D l lin N on irec W g -R tin rit ea g i di (re Gr ng ng a am di R ng m a el at re r ed la t In ed st ) ru F cti ee o d n U bac nc k od ab le B Percent Time oo k an d O ra l P L Pr a ng in ua t ho A ge w ne ar m en ic es Aw s Le ar tte en rR es A lp s ha e c b e og n tic iti on In S st tru r uc ct ur tio al n A na ly si W or s d W or V k oc P re ab vi u ew la ry in S g pe a lli B ng oo in k R C ea R on ea di te ng di xt Th ng B R oo e a e ir O ks di w ng n W C om rit in pr eh g en si on S pe lli ng D W i r rit N o n ect in in g -R G g r ea (r am ea di m ng di ar ng R el re at ed late d) In st ru ct i Fe o n ed ba c U nc k od ab le B Percent Time Time spent in Reading/LA Activities in 1st grade by Hi vs. Low Rated Implementers 0.18 0.2 0.16 Hi 0.14 Low 0.12 0.08 0.1 0.06 0.04 0.02 0 Time spent in Reading/LA Activities in 2nd Grade by Hi vs. Low Rated Implementers 0.18 0.2 0.16 Hi 0.14 Low 0.12 0.08 0.1 0.06 0.04 0.02 0 A Hypothetical Model of How Teacher Variables Moderate the Impact of Student’s Initial Reading Ability on Reading and Spelling Outcomes Growth in Total Reading Skill Before, During, and Following Intensive Intervention (Torgesen et al., 2001) 95 Standard Score 90 85 LPSP EP 80 75 P-Pretest Pre Post 1 year 2 year Interval in Months Between Measurements Time x Activity Analyses for the Two Intervention Approaches LIPS EP Phonemic Awareness and Phonemic Decoding 85% 20% Sight Word Instruction 10% 30% Reading or writing connected text 5% 50% Reading rate remained quite impaired 100 Accuracy-91 90 80 Rate-72 70 Pretest Posttest 1-year 2-year Remediation is not a solution! Reading rate is limited because the proportion of words in grade level passages that children can read “by sight” is less than for average readers. How do you close the gap when the student is already 3- 5 years behind? Yet, there are some impressive results • Berninger et al., 2003; Blachman et al., 2004; Olson & Wise, 2006 • Lovett et al. (2000): PHAB/DI + WIST → PHAST Track Reading Program • Wolf, Miller, & Donnelly’s (2002) RAVE-O Effective Early Interventions • Reading Recovery: Schwartz’s (2005) RCT concludes that 5% of RR graduates don’t read on grade level. • Peer Assisted Learning Strategies (PALS): Studies show that 5-6% of 1st graders read above 30th %ile. • Mathes et al. (RRQ; 2005) Peer Assisted Learning Strategies • As a supplement to core reading, PALS has helped K-6 graders improve their phonological awareness, phonics, fluency, and comprehension (e.g., Fuchs et al., 1997; Mathes et al., 1994; Mathes et al., 1998; Simmons et al., 1994). • Teachers pair their students, creating dyads with one high and one low performing reader, and then train students to follow standard PALS procedures. Increases students’ practice time and opportunities to respond. Offers structured and reciprocal practice on phonological awareness, phonics, fluency, and comprehension Mathes et al. (2005) Children – sampled across 2 years • 300 At-Risk Readers identified with the Texas Primary Reading Inventory - assigned randomly to intervention. • 100 Typically Developing Readers Teachers • 6 Intervention (3 Proactive & 3 Responsive) • 30 General Education 1st-grade Teachers Schools • 6 non- Title 1 elementary schools in a large urban school district with an aggressive, long- term reading initiative The Interventions Enhanced Classroom Instruction All children identified as at-risk by principal, teachers, and parents Progress monitored with feedback to principal, teachers, and parents (oral reading probes every 3 weeks) Professional development of classroom teachers in strategies for accommodating academic diversity and linking assessment to instructional planning for struggling readers Comparison of Two Interventions Proactive and Responsive • 40 minutes, 5 days per week, all school year (30 weeks) • 1:3 teacher-student ratio • Taught by certified teachers who are school employees, but trained and supervised by researchers • Provided in addition to enhanced classroom instruction Proactive Intervention • Explicit instruction in synthetic phonics, with emphasis on fluency. • Integrates decoding, fluency, and comprehension strategies. • 100% decodable text. • Carefully constructed scope and sequence designed to prevent possible confusions. • Every activity taught to 100% mastery everyday. Responsive Intervention Explicit instruction in synthetic phonics and in analogy phonics. Teaches decoding, using the alphabetic principle, fluency, and comprehension strategies in the context of reading and writing. No pre-determined scope and sequence. Teachers respond to student needs as they are observed. Leveled text not phonetically decodable. The Responsive Intervention • Fluency Work (Repeated Reading) and Assessment: 8-10 minutes • Word Work: 10-12 Minutes • Supported Reading: 10-12 Minutes • Supported Writing: 8-10 Minutes Predicted Growth in Word Reading by Group - Year 1 & 2 1.5 1 Z-score 0.5 0 Low Risk Responsive Classroom Proactive -0.5 -1 -1.5 October December February Month April Predicted growth in CMERS by group 100 90 Low Risk Responsive Classroom Proactive 80 Raw Score 70 60 50 40 30 20 10 0 1 2 3 4 5 6 Probe 7 8 9 10 11 Reading Outcomes Across Critical Domains 115 110 105 Classroom 100 Proactive 95 Responsive Benchmark Not At-Risk 90 85 80 Word Rec Fluency Comprehension At Risk Reader Left Right Kindergarten First Grade Simos et al., 2006 What percent of children don’t respond adequately to quality intervention? Primary only: 15/92 = 16% (3% of school population) Primary + Secondary: Proactive: 1/80 = < 1% (< .2% of school population) Responsive: 6/83 = 7% (<1.5% of school population) Denton, Fletcher, Anthony, & Francis (2006; JLD) Wave 1 Round 1 Pre Wave 2 Phono-Graphix P 8 weeks Round 2 Pre Baseline 8 weeks Wave 3 Read Naturally 8 weeks Pre RN Phono-Graphix 8 weeks P Wave 4 8 weeks Read NaturallyRN 8 weeks Gains in Basic Skills Standard Score Points During 16-Week Intervention 30 Standard Score Gains 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -5 Students 16 17 18 19 20 21 22 23 24 25 26 27 Conclusions • Significant improvements in decoding, fluency, and comprehension after 8 weeks of PhonoGraphix. Small to moderate effects of Read Naturally on fluency only, perhaps due to need for more decoding before repeated reading. • 7 of the 27 students performed at or above the 30th %ile of the WJ-III Basic Reading after 16 weeks of daily 2-hr.intervention (& 4 between the 25th and 30th). • Nonresponders with Tier 1 + 2 > Tier 1 alone. • Development of reading skills dependent on establishment of LH neural network (Simos et al., 2007, JLD) Are Volunteer Tutors Effective? • Overall mean effect size for tutoring in several large meta-analyses is .40 (Cohen et al., 1982; Elbaum et al., 2000). • Average effect size for volunteers was .26; however, in studies describing tutors’ training, effect size was .59 (Elbaum et al.) Effect sizes on components of reading • Word identification: .42 (Baker et al., 2000) to 1.24 (Invernizzi et al., 1997) • Word attack: .32 (Vadasy et al., 1997) to 1.24 (Vadasy et al., 2000) • Fluency: .48 (Baker et al., 2000) to .53 (Baker et al., 2000) • Comprehension: .10 (Vadasy et al., 2002) to .32 (Baker et al., 2000), .90 (Al Otaiba et al., 2005) Support for community tutors (Wazik, 1998) • Certified reading specialist to supervise tutors • Ongoing training and feedback for tutors • Structured tutoring sessions that incorporate basic literacy elements • Consistent/intensive tutoring for struggling readers • Access to high quality materials • Ongoing assessment of student progress • Monitoring of attendance • Coordination of tutoring with classroom instruction Multi- Tiered Reading Instruction If progress is inadequate, move to next level. Level 1: Primary Intervention Enhanced general education classroom instruction (90 min, uninterrupted). Level 2: Secondary Intervention Child receives more intense instruction in general education in small groups (30 min). Level 3: Tertiary Intervention increases in intensity and duration; remedial, small groups (30+ min.) Who is LD, What is RtI? • The student who does not respond to quality instruction • Discrepancy relative to the expectation that ALL children can learn • Requires closer integration of general education and special education • One system, not two -- all students are general education students first! IDEA 2004 • NCLB and IDEA share the goal of a single, wellintegrated system that connects general, remedial, and special education and considers the learning needs of all children. • According to IDEIA, response to intervention (RTI) means that a local education agency “may use a process that determines if the child responds to scientific, research-based intervention as a part of the evaluation procedures” (Pub. L. No. 108-446 § 614 [b][A]; § 614 [b] [2 & 3]). RtI as a Diagnostic System vs. Multi-Tiered Instructional Model • It is useful to keep RtI as a diagnostic system conceptually distinct from RtI as a multi-tiered instructional model because the former is new and has challenging measurement implications, whereas the latter has been in existence in public health and in school reform models. What is RtI? RTI is the practice of (1) providing high-quality instruction/intervention matched to student needs and (2) using learning rate over time and level of performance to (3) make important educational decisions to guide instruction. National Association of State Directors of Special Education, 2005 REFERRAL SCREENING ELIGIBILITY TESTING Not Eligible Eligible TREATMENT Responders Non-Responders NEW MODEL TREATMENT 1-2 Responders Monitor Non-Responders ELIGIBILITY TESTING Not Eligible Eligible TREATMENT 3 Responders Non-Responders Monitor Three Tier Model • Critical question for special education and general education is not “What is the label?” or “For what is the child eligible, but what type of intervention results in change? Special Education becomes an alternative for students who don’t respond to quality instruction and need the power/flexibility of IDEA Implementing 3 tier models • Enhanced core reading instruction is the key • Primary model: begins in the classroom with professional development, assessment, and better materials; alternatives like PALS underutilized • Screening, diagnostic assessments, and progress monitoring must be in place • Goal is differentiated instruction and monitoring response to instruction Growth in Oral Reading Fluency for STACY,JUSTIN 3 120 85 WPM 100 Score 80 60 40 20 0 Story 1 Story Story 2 3 Story Story 4 5 Story Story 6 7 Story 8 Story Story 9 10 School Year 21 weeks Story Story 11 12 Story Story 13 14 Story 15 Implementing 3 Tier models • Second tier is typically small group pull out instruction, but can represent additional dose in the classroom • Small-group intervention is just as effective as 1:1 intervention (Elbaum et al., 2000) • In reading, content is the same as for effective classroom intervention: explicit instruction in the alphabetic principle, reading for meaning and opportunities to learn (Foorman & Torgesen, 2001) Implementing 3 Tier models • Third tier is typically special education, but can represent more intense tutoring in general education • Content significantly different from first 2 levels • Response to Instruction (RtI) should be part of the criteria for determining eligibility for special education • Should be measured, not surmised • Progress monitoring essential Reading Improvement is a Systemic Undertaking STUDENT TEACHER CONTENT PELP Coherence Framework The Kennewick, WA Success Story: Reading Improvement Requires… • Data: good assessments—benchmark and normative—and expert use of the data • Increased direct instructional time; additional time for those behind • Quality instruction in small, fluid, skill groups • Targeted accelerated growth; knowledgeable reading specialists Fielding, Kerr, Rosier, 2007 Instructional leadership at Kennewick • Instructional conferences for all administrators (viewing videotaped lessons) • Learning walks (to observe lesson purpose and rigor and student engagement; debrief) • The two-ten goal (administrators spend 2 hrs/day or 10 hrs/week on instructionally focused activities) • Literacy coaches at middle and high school (meet weekly with principal to plan instruction & PD; confer regularly with teachers) Initial status + Growth = Outcome • Correlation of initial achievement and ending achievement is .83-.90. • Students who start ahead, stay ahead; students who start behind, stay behind. • Schools don’t create the achievement gap; they inherit it. 13 higherSES children (professional) 23 middle/lowerSES children (working class) 6 welfare children Age of child in months Hart & Risley, 1995 Estimated cumulative words addressed to child Language Experience Professional Working-class Welfare Age of child in months Hart & Risley, 1995 Table 3 % 98 90 80 70 60 50 40 30 20 10 2 Independent Reading Minutes Per Day 65.0 21.1 14.2 9.6 6.5 4.6 3.3 1.3 0.7 0.1 0.0 Words Read Per Year 4,358,000 1,823,000 1,146,000 622,000 432,000 282,000 200,000 106,000 21,000 8,000 0 Variation in Amount of Independent Reading (Cunningham & Stanovich, 1998, adapted from Anderson, Wilson, & Fielding,1988) Early Learning is Crucial • Narrowing the achievement gap before kindergarten is a powerful, proactive, and doable task. • Build oral language and literacy development into pre-K classes • Have parents read to their children 20 min. a day to expose them to rare vocabulary, complex syntax, and rich discussion. For more information…. Foorman, B. R., & Al Otaiba, S. (in press). Reading Remediation: State of the Art. In K. Pugh and P. McCardle (Eds.), How children learn to read: Current issues and new directions in the integration of cognition, neurobiology and genetics of reading and dyslexia research and practice. San Antonio, TX: Pro-Ed. Go to www.FCRR.org BFoorman@fcrr.org Thank you!