Term
| Traditional Model of SLD Identification |
|
Definition
Focuses on within-child ability/achievement discrepancy
•Causes of SLD presumed to be due to internal/organismic variables
•Unexpected underachievement relative to ability
•IQ-achievement discrepancy as a marker for presence of SLD
•Assumes better classification leads to better treatment (finds the right kids)
•Presumes stability of ability & achievement over time |
|
|
Term
|
Definition
Franzen (1920) & Accomplishment Quotient
–AQ = Educational Quotient/Intelligence Quotient (AQ=EQ/IQ)
–IQ seen as limiting factor in educational accomplishment
–“Mild & jug” conceptualization (can’t put a gallon milk in a quart jug)
–IQ>EQ=Underachievers
–IQ<EQ=Overachievers |
|
|
Term
| Grade Level Discrepancy (origins of discrepancy) |
|
Definition
–Expected grade level-Actual grade level (EGL-AGL)
–Example: EGL=5 AGL=5 AGL-EGL=0 (on grade level)
-Example: EGL=5 AGL=3 AGL-EGL=5-2=3 grade levels below |
|
|
Term
| Standard Score Difference Methods |
|
Definition
Origins of Discrepency
–IQ-Achievement (100-80=20 point discrepancy)
–Discrepant versus nondiscrepant underachievers |
|
|
Term
| Regression-Based Discrepancy Methods |
|
Definition
-Trying to predict achievement from IQ
–Corrects for regression effects
–Recall the SEM
–Consider the SEE (standard error of estimate)
-takes into account the correlation of 2 scales
–Considerations
•Reliability of IQ test (always < 1)
•Reliability of achievement test (always < 1)
•Correlation between IQ & achievement tests (always < 1)
•Reliability of difference scores (always < reliabilities of IQ & achievement tests) |
|
|
Term
| Evaluation of Regression Methods |
|
Definition
–Considers regression effects
–Discrepancies are test specific (psychometric true score versus absolute true score)
–Statistically significant discrepancies versus severe discrepancies
–How much discrepancy required to differentiate low achievement from SLD?
–Is SLD “real” and different from garden variety low achievement? |
|
|
Term
|
Definition
–Reviewed 19 studies contrasting discrepancy & nondiscrepant low achievers
–Few differences on measures of cognitive skills related to reading
•Psuedoword reading
•Real word phonetic analysis
•Speech related phonological processing
•Automaticity
•Spelling
•BOTH GROUPS SHARE A GENERAL PHONOLOGICAL PROCESSING DEFICIT |
|
|
Term
|
Definition
| Focus on & manipulation of phonemes in spoken words |
|
|
Term
|
Definition
–Focus on letter-sound correspondence of words in print
•s/t/o/p or p/l/a/y
•Synthetic phonics
•Embedded phonics
•Analytic phonics
•Onset-rime phonics |
|
|
Term
|
Definition
–46 studies contrasting discrepant & nondiscrepant low achievers
–Contrasted groups in domains of achievement, behavior, & cognitive skills
–Found few differences between discrepant & nondiscrepant low achievers
–Concluded that IQ-achievement had little evidence for validity |
|
|
Term
Role of Intelligence in Identification of SLD
meta-analyses |
|
Definition
–Steubing et al. (2002) found few differences between discrepant/nondiscrepant LAs
–Steubing et al (2006) found that IQ did not predict RTI (R2=0.3%)
–Hoskyn & Swanson (2000) found few differences between discrepant/nondiscrepant LAs
–Fuchs & Young (2006) found 8 of 13 studies showed IQ predicted RTI (“scorecard” analysis) |
|
|
Term
| Vellutino et al. (1996) Study |
|
Definition
•55-year longitudinal study from Kg-4th grade
•2 groups identified
–Struggling readers (n=125) at or below 15%ile on WRMT (Word Attack/Word Identification)
–Normally developing readers (n=65) at or above 40th %ile on WRMT
•IQ 90 and above on WISC-R
•Intervention
–Mid first grade
–Daily one-one tutoring 30 minutes per day 5 days per week for 15 weeks (35 hours)
–After one semester, all children rank ordered in terms of growth in reading
•4 growth categories
–Very Limited Growth (VLG)
–Limited Growth (LG)
–Good Growth (GG)
- Very Good Growth (VGG) |
|
|
Term
| Vellutino et al. Findings |
|
Definition
•4 tutored groups did not differ on IQ & did not differ from normal readers in IQ
•Difficult-to-remediate children differed greatly on phonological skills
•No strong reliable relationships between IQ & reading achievement
–Word Identification: 0.06-0.16 from Grades 1-4
–Word Attack: -0.02-0.05 from Grades 1-4
•Suggests that RTI may be more valid means of identifying SLD
•Hierarchical Multiple Regression Analyses
–FSIQ-Word Attack-Word Identification-Basic Skills
–IQ accounted for 0% of variance across reading skills measures
–Word Attack-Word ID-Basic Skills predicted large % of variance in subsequent reading
–36%-73% (R2 )
**Best predictor of future reading is past reading |
|
|
Term
| Velluntino summary of conclusions |
|
Definition
-IQ accounts for a small amount of unique variance in predicting intervention response.
-IQ-discrepancies are weakly related to achievement and cognitive differences relative to simple low achievement and to prognosis and they present significant psychometric
problems. |
|
|
Term
WISC-IV
(areas of strengths/weaknesses) |
|
Definition
–Verbal Comprehension
–Perceptual Reasoning
–Working Memory
–Processing Speed |
|
|
Term
Stanford-Binet V
(areas of strengths/weaknesses) |
|
Definition
–Fluid Reasoning (verbal & nonverbal)
–Knowledge (verbal & nonverbal)
–Quantitative Reasoning (verbal & nonverbal)
–Visual-Spatial Processing (verbal & nonverbal)
–Working Memory (verbal & nonverbal) |
|
|
Term
Woodcock-Johnson Cognitive-III (CHC theory)
(Areas of strengths/weaknesses) |
|
Definition
–Crystallized intelligence
–Fluid intelligence
–Auditory intelligence
–Visual intelligence
–Spatial intelligence
–Working Memory |
|
|
Term
Kaufman Assessment Battery for Children-II
(areas of strengths/weaknesses) |
|
Definition
–Learning (long-term retrieval)
–Sequential (short-term memory)
–Simultaneous (visual)
–Planning (fluid)
–Knowledge (crystallized)
–Fluid-Crystallized Index |
|
|
Term
Cognitive Assessment System
(areas of strengths/weaknesses) |
|
Definition
–Planning
–Attention
–Simultaneous
–Successive |
|
|
Term
Differential Ability Scales
(areas of strengths/weaknesses) |
|
Definition
–Verbal Ability
–Nonverbal Ability
–Spatial Ability |
|
|
Term
|
Definition
•Group mean differences & diagnostic accuracy
•ACID profile on WISC (Arithmetic-Coding-Information-Digit Span)
•SLD students show lower ACID scores than non-SLD students
•Probability of SLD child showing ACID profile is high given that child is SLD
•Not how referrals take place in schools
–We want to know the probability that student is SLD given an ACID profile
–Known as inverse probability which overestimates diagnostic accuracy
**inverse Bayesian probability |
|
|
Term
| Group mean differences not diagnostically accuracy |
|
Definition
–85% of teachers are women
–2% of women are teachers
–43% of men on death row are black
–<1% of black men are on death row
–SLD students show an ACID profile more often than non-SLD students
–Given an ACID profile, how many students are SLD? |
|
|
Term
| Learning Disabilities Association of America (5 main conclusions) |
|
Definition
•Survey of “Expert Panel” (n=56) professionals
•5 Major Conclusions
–Maintain SLD definition & strengthen statutory requirements for SLD identification
–Neither IQ-achievement discrepancy nor RTI alone is sufficient for SLD identification
–“3rd Method” focuses on assessment of processing strengths/weaknesses makes most empirical/clinical sense
–Comprehensive evaluations required for SLD identification & to distinguish from LA
–PSW assessment used for both identification & intervention purposes |
|
|
Term
LDA white paper:
Maintain current SLD statutory definition |
|
Definition
–PSW hallmark of SLD
–Children with LA should be excluded from definition
–SLD qualitatively & functionally different from LA
–Chronically nonresponsive children need something different & individualized instruction |
|
|
Term
LDA white paper:
Neither IQ-achievement discrepancy nor RTI sufficient for SLD identification |
|
Definition
–Major problems with discrepancy approaches
–RTI measures lack technical adequacy for SLD decision making
–No consensus on type of RTI (standard protocol or problem solving)
–Single case designs cannot be used because manipulation of more than 1 IV not possible
–No true positive in RTI model (SLD by default) |
|
|
Term
LDA white paper:
"3rd Method” Needed based on PSW |
|
Definition
–Recognizes statutory requirements
–Recognizes relationship between PSW & academic achievement deficits
–Existence of technically adequate PSW assessment tools |
|
|
Term
LDA white paper:
Comprehensive evaluations of PSW needed to distinguish SLD from LA |
|
Definition
–SLD children need individualized interventions not merely more intense interventions
–Comprehensive evaluations required as early as possible (Tier 2)
–SLD children fundamentally “different” from LA children |
|
|
Term
LDA white paper:
PSW assessment used for both identification & intervention purposes |
|
Definition
–Research only recently begun to emerge
–More evidence in reading than other areas (math, written language)
–Further research needed |
|
|
Term
| Consortium for Evidence-Based Early Intervention Practices |
|
Definition
•117 citations used in White Paper fails to support 5 conclusions
•85 of 117 citations (73%) are commentary articles, book chapters, or books often authored by experts on panel
•15/117 articles are correlational studies (no mediation studies)
•100/117 (85%) of studies show no causal relation between PSW & achievement
•Comprehensive assessment of PSW not required in IDEIA
•No data support PSW can be used to improve treatment outcomes
•No qualitative criteria to indicate a person “has” or “is” SLD
•No unique pattern of PSW of either adequate or inadequate responders
•Psychometric problems with PSW discrepancy scores |
|
|
Term
| Psychometric problems with PSW discrepancy scores |
|
Definition
–10 subtests produces 45 combinations of potential PSWs
–20 subtests produces 90 combinations of PSWs
–Cross-battery assessments produces geometrically astronomical combinations
–Problems with reliability of difference scores |
|
|
Term
| Processing Approaches in SLD Identification |
|
Definition
•Some argue that law requires processing assessment for SLD
•Noting written in law requires documentation of processing disorder
•37 states and federal regulations do not require processing assessment
•Only 13 states mention processing but do not address how it is established
•IDEIA (2004) |
|
|
Term
| take home message on processing approaches |
|
Definition
•There is not a single randomized clinical trial using IES evidence-based standards that has related processing strengths to effective intervention outcomes
•Relations between processing & outcomes are correlational, not causal |
|
|
Term
|
Definition
•Moderator variable (B) affects direction or strength between predictor (A) and criterion (C)
•Impact of A on C varies with level or value of B (correlational)
•Mediator specifies manner or mechanism by which an effect occurs (causal)
|
|
|
Term
| correlations vs. causation |
|
Definition
•Consider the following:
–Verbal Comprehension (VC) and Working Memory (WM) relation to Reading Deficit (RD)
–VC correlates with RD
–WM correlates with RD
–VC correlates with WM
•Following is true:
–Poor readers have lower VC than average or good readers
–Poor readers have lower WM scores than average or good readers
–Poor readers have higher rates RD than average or good readers
•Following is not logically true:
–VC and WM deficits cause RD (logical fallacy of affirming the consequent) |
|
|
