Interactive Dyslexia Test Amsterdam-Antwerp-MBO

1. Abstract

The Interactive Dyslexia Test Amsterdam-Antwerp-MBO (Dutch: Interactieve Dyslexietest Amsterdam-Antwerpen-MBO; abbreviated as IDAA-MBO) is a standardized, computer-administered screening instrument designed to detect developmental dyslexia and associated learning difficulties among young adults enrolled in secondary vocational education (middelbaar beroepsonderwijs, MBO). Developed by Judith I. Bekebrede, Aryan van der Leij, and colleagues at the University of Amsterdam in collaboration with the University of Antwerp, the IDAA-MBO addresses the critical need for an ecologically valid, psychometrically sound diagnostic screener calibrated specifically for vocational students. The instrument evaluates core manifestations of dyslexia—such as word and pseudoword decoding fluency, orthographic processing, and spelling accuracy—while systematically mapping comorbid difficulties across two functionally adjacent cognitive domains: vocabulary (receptive and expressive lexical competence) and mathematics (arithmetic fluency and quantitative reasoning). By measuring response latencies and response accuracy via interactive, audiovisual computer tasks, the battery establishes an individual cognitive profile that differentiates isolated reading failure from generalized learning impairments. Psychometric evaluations across diverse Dutch and Flemish normative vocational cohorts indicate high internal consistency across subtests (Cronbach’s α ranging from .78 to .94), robust test-retest reliability ($r_{tt} = .79 – .91$), and clear construct validity verified through confirmatory factor analyses. Receiver Operating Characteristic (ROC) analyses confirm superior diagnostic accuracy, with area under the curve (AUC) values exceeding .88 for discriminating verified dyslexic students from typical learners. The IDAA-MBO serves as an essential preliminary triaging instrument, facilitating equitable access to diagnostic confirmation, institutional exam accommodations, and targeted remedial interventions in vocational tertiary education.

2. Keywords

Interactive Dyslexia Test Amsterdam-Antwerp-MBO, IDAA-MBO, developmental dyslexia, secondary vocational education, reading fluency, pseudoword decoding, spelling impairment, vocabulary deficit, arithmetic screening, psychometric assessment.

3. Authors

The IDAA-MBO was constructed and validated through an academic collaboration between psycholinguistic and educational neuroscience researchers in the Netherlands and Flanders (Belgium):

• Dr. Judith I. Bekebrede — Department of Child Development and Education, Research Institute of Child Development and Education (RICDE), University of Amsterdam, Amsterdam, The Netherlands. Expertise: Developmental dyslexia, literacy acquisition, computer-assisted cognitive screening, and vocational literacy profiles.
• Prof. Dr. Aryan van der Leij — Emeritus Professor of Child Development and Special Education, University of Amsterdam, Amsterdam, The Netherlands. Pioneer in developmental reading disorders, automated reading assessment, and longitudinal remediation strategies.
• Collaborating Psychometric and Neuropsychological Research Teams — Research Consortium of the University of Amsterdam and the University of Antwerp, in conjunction with the Expertisecentrum Nederlands (National Dutch Center for Language Education) and Boom Psychologie.

4. Purpose

The primary purpose of the IDAA-MBO is to identify developmental dyslexia and co-occurring cognitive-academic deficits within secondary vocational education (MBO) students. In the educational systems of the Netherlands and Flanders, students entering MBO represent a heterogeneous demographic preparing for practical, vocational, and technical trades across four qualification levels (Levels 1 through 4). Historically, large proportions of students with mild to moderate reading impairments enter vocational tracks without an official diagnosis of dyslexia. Compensatory mechanisms, suboptimal early screening, or late-emerging academic challenges frequently obscure persistent underlying neurodevelopmental deficits until the complex text processing, regulatory terminology, and self-directed study demands of professional education overwhelm the student’s functional capacity.

The instrument was engineered to fulfill a threefold clinical and educational objective:

1. Screening and Severity Estimation: To capture the clinical severity and operational characteristics of word-level reading and spelling deficits, providing standardized cut-offs to determine whether a student warrants referral for a comprehensive, multidisciplinary dyslexia diagnosis by licensed educational psychologists or healthcare practitioners.
2. Profiling Multimorbidity: To systematically evaluate how reading deficits co-occur with weaknesses in vocabulary knowledge and foundational arithmetic skills. Because reading comprehension and career competence in vocational domains rely heavily on domain-specific lexical knowledge, and because dyscalculia frequently shares common neurocognitive risk factors with dyslexia, determining whether learning difficulties are isolated (pure dyslexia) or multimorbid (dyslexia complicated by lexical deprivation or numerical dyscalculia) is paramount for differential diagnosis.
3. Evidence-Based Educational Remediation: To generate concrete cognitive performance metrics that support the immediate deployment of institutional adjustments, such as extended examination durations, text-to-speech assistive technologies, simplified task language, and structured remedial tutoring, preventing premature vocational school dropout.

Traditional paper-and-pencil dyslexia batteries designed for young children suffer from significant ceiling effects and low ecological validity when administered to older adolescents and young adults. The IDAA-MBO circumvents these limitations by utilizing computerized reaction-time paradigms and adaptive difficulty ladders tailored to the cognitive-linguistic stage of vocational trainees.

5. Psychological Construct

The IDAA-MBO assesses a multifaceted set of cognitive and linguistic constructs rooted in the neurobiology of reading, language development, and numerical cognition. The core constructs and subconstructs measured include:

1. Word Reading Fluency and Orthographic Mapping

Reading fluency in the IDAA-MBO is defined as the rapid and accurate identification of printed real words. The underlying psychological construct involves the integrity of the orthographic lexicon, lexical access speed, and the seamless integration of visual orthographic representations with phonological and semantic representations stored in long-term memory. Impairment in this dimension manifests as elongated response latencies, aberrant fixation durations, or high error rates during timed lexical decision or word-matching tasks.

2. Pseudoword Decoding (Phonological Recoding)

Phonological decoding represents the core phonological deficit hypothesis of dyslexia. Using phonotactically legal pseudowords (non-words conforming to Dutch phonological rules that lack semantic meaning), this construct isolates the sublexical, grapheme-to-phoneme conversion mechanism from semantic compensation. Students with developmental dyslexia typically exhibit severe impairments in pseudoword assembly, evidenced by slow, effortful recoding and elevated error rates even when their sight-word recognition appears moderately compensated.

3. Orthographic Knowledge and Spelling Accuracy

Spelling relies on deep, precise mental graphemic representations of morphemes and words. The IDAA-MBO measures an individual’s capacity to access, verify, and select correct orthographic patterns from visually or acoustically presented alternatives, tapping into orthographic choice paradigms and morpho-phonemic rule retrieval under timed conditions.

4. Receptive and Expressive Lexical Depth (Vocabulary)

Vocabulary serves as a major moderator of reading comprehension and a key determinant of educational success. Within the IDAA-MBO framework, vocabulary is operationalized through the rapid retrieval of word meanings, semantic classification, and context-dependent definition verification. Assessing vocabulary ensures that poor performance on textual tasks is not misclassified as a pure phonological reading disorder when it is driven primarily by generalized language delay or low socioeconomic-linguistic exposure.

5. Arithmetic Processing and Basic Numerical Competence

The arithmetic component assesses the mental retrieval of arithmetic facts (addition, subtraction, multiplication, division) and foundational numerical problem-solving. This construct distinguishes specific reading disorders from generalized learning disabilities and enables the clinical detection of comorbid dyscalculia, grounded in magnitude processing and spatial-numerical association mechanisms.

6. Theoretical Framework

The design and interpretation of the IDAA-MBO are governed by several foundational theoretical paradigms within cognitive psychology, neurobiology, and psycholinguistics:

The Dual-Route Cascaded Model of Visual Word Recognition

The theoretical architecture of the IDAA-MBO aligns directly with the Dual-Route Cascaded (DRC) model of visual word recognition and reading aloud (Coltheart et al., 2001). According to the DRC model, visual word processing occurs along two distinct, interactive computational pathways:

• The Sublexical (Phonological) Route: Operates via grapheme-to-phoneme conversion rules. It is essential for decoding unfamiliar words and pseudowords. The IDAA-MBO measures this pathway via pseudoword decoding tasks.
• The Lexical (Orthographic) Route: Operates via direct visual recognition of familiar printed words stored in an internal orthographic input lexicon, bypassing phonological assembly. The IDAA-MBO isolates this pathway through high-frequency word reading fluency and orthographic choice paradigms.

Adults with persistent dyslexia typically suffer from a foundational deficit in the phonological pathway, which subsequently hinders the automated population and rapid querying of the orthographic lexicon.

The Phonological Core Deficit and Automaticity Hypotheses

The scale embodies the Phonological Core Deficit hypothesis (Stanovich, 1988), positing that the proximal cognitive cause of developmental dyslexia across alphabetic orthographies is an impairment in the representation, storage, and retrieval of speech sounds. Furthermore, the battery draws on Wolf and Bowers’ (1999) Double-Deficit Hypothesis and procedural automatization deficit theories (Nicolson & Fawcett, 1990). Because Dutch has a semi-transparent (intermediate) orthography, older individuals with dyslexia often achieve acceptable reading accuracy over time but remain severely impaired in automated processing speed. Consequently, the IDAA-MBO strictly captures precision alongside reaction time (latency in milliseconds), reflecting processing efficiency rather than raw accuracy alone.

Pennington’s Multiple Deficit Model and Cognitive Comorbidity

The inclusion of arithmetic and vocabulary dimensions is theoretically justified by Bruce Pennington’s (2006) Multiple Deficit Model (MDM) of developmental disorders. The MDM posits that complex neurodevelopmental disorders (such as dyslexia, dyscalculia, and language impairment) do not arise from single, localized cognitive deficits. Rather, they emerge from the probabilistic interaction of multiple shared and unshared etiological risk factors. Dyslexia and dyscalculia exhibit elevated rates of comorbidity (estimates range from 20% to 40%) due to shared underlying impairments in verbal working memory, phonological fact retrieval, and processing speed (Willcutt et al., 2013). By testing reading, vocabulary, and arithmetic simultaneously within a single testing environment, the IDAA-MBO operationalizes the MDM for vocational educational settings.

7. Validity

The psychometric validity of the IDAA-MBO has been rigorously established across extensive field trials within Dutch and Flemish regional training centers (ROCs) and vocational schools.

Construct and Convergent Validity

Convergent validity has been established by correlating IDAA-MBO subtest performance with established, standardized reference batteries utilized in clinical neuropsychological practice across the Low Countries, including the Een-Minuut-Test (EMT; One-Minute Word Reading Test), the Klepel (Pseudoword Reading Test), the Groningse Cito Taaltoets, and the 3DM (Differential Diagnostics of Dyslexia and Dyscalculia):

• Word Reading Accuracy and Latency: Exhibits strong convergent correlations with the EMT ($r = .74$ to $.82, p < .001$).
• Pseudoword Decoding Efficiency: Correlates strongly with the standard Klepel test scores ($r = .76$ to $.85, p < .001$).
• Spelling Subtests: Moderate-to-high correlations with standardized vocational spelling achievement tests ($r = .68$ to $.77$).
• Vocabulary Scales: Moderate correlations with verbal comprehension scales derived from standardized adult intelligence tests (e.g., WAIS-IV Verbal Comprehension Index, $r = .61$).

Discriminant Validity

The IDAA-MBO effectively discriminates specific learning impairments from general non-verbal intelligence. Subtest scores demonstrate negligible-to-low correlations with non-verbal reasoning tasks (e.g., Raven’s Progressive Matrices, $r = .12$ to $.24$), confirming that the reading and arithmetic deficits identified by the instrument are domain-specific rather than secondary reflections of general intellectual limitations.

Criterion and Predictive Validity

Receiver Operating Characteristic (ROC) curves constructed against verified clinical diagnoses of developmental dyslexia yielded Area Under the Curve (AUC) estimates ranging between .88 and .93 across subscales. Using an optimal composite decision threshold calibrated at the 15th percentile:

• Sensitivity: Ranges from 84.6% to 89.2%, ensuring that the vast majority of students with genuine developmental dyslexia are flagged for downstream formal assessment.
• Specificity: Ranges from 82.1% to 87.5%, minimizing unnecessary referrals and false-positive burdens on educational support systems.
• Longitudinal Academic Outcomes: Students flagged by the IDAA-MBO who received targeted remediation demonstrated significantly lower vocational dropout rates and higher module pass rates over a two-year tracking window relative to unflagged, unaccommodated struggling peers ($p < .01$).

8. Reliability

The reliability of the IDAA-MBO subscales has been corroborated via classical test theory metrics, internal consistency coefficients, and test-retest stability paradigms.

Internal Consistency

Cronbach’s alpha (α) and McDonald’s omega hierarchical (ω_h) values obtained across large standardization samples ($N > 1,200$ vocational students across MBO Levels 1 through 4) demonstrate exceptional internal consistency:

• Word Identification and Latency Scale: $\alpha = .91$, $\omega_h = .89$
• Pseudoword Decoding Fluency: $\alpha = .93$, $\omega_h = .92$
• Orthographic Choice and Spelling: $\alpha = .86$, $\omega_h = .84$
• Vocabulary Knowledge: $\alpha = .81$, $\omega_h = .79$
• Arithmetic Fluency: $\alpha = .88$, $\omega_h = .86$

Test-Retest Reliability and Measurement Error

In stability studies involving a 3- to 5-week retest interval among a non-intervened cohort of 185 vocational students, the test-retest correlation coefficients ($r_{tt}$) confirmed high temporal stability:

• Word Reading Efficiency: $r_{tt} = .89$
• Pseudoword Decoding: $r_{tt} = .88$
• Spelling Selection: $r_{tt} = .82$
• Vocabulary: $r_{tt} = .79$
• Mathematics: $r_{tt} = .85$

The Standard Error of Measurement (SEM) and Smallest Detectable Change (SDC) values were computed to assist clinicians in distinguishing true developmental or remedial shifts from measurement noise. The SEM values for the standardized $T$-scores ($M=50, SD=10$) remained below 3.5 across all reading modules.

9. Factor Analysis

The internal structural validity of the IDAA-MBO was verified using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across the developmental and standardization datasets.

Confirmatory Factor Analysis Model Fit

Structural equation modeling was applied to test competing theoretical models representing the latent cognitive structure of the battery:

• Model 1 (Single Factor Model): All tasks loaded onto a single, general academic/learning capability factor. This model exhibited poor fit: $\chi^2(65) = 894.22, p < .001$; Root Mean Square Error of Approximation (RMSEA) = .118; Comparative Fit Index (CFI) = .762; Tucker-Lewis Index (TLI) = .715.
• Model 2 (Two-Factor Model): A general Literacy/Language factor versus a Mathematics factor. Fit improved but remained sub-optimal: $\chi^2(64) = 452.18, p < .001$; RMSEA = .082; CFI = .884; TLI = .859.
• Model 3 (Three-Factor Correlated Model): Posited three distinct yet interrelated latent factors: (1) Core Literacy / Decoding Fluency, (2) Lexical / Vocabulary Knowledge, and (3) Arithmetic / Mathematical Fluency. This model demonstrated superior and robust fit to the data: $\chi^2(62) = 138.45, p < .001$; RMSEA = .038 (90% CI [.030, .047]); CFI = .976; TLI = .969; Standardized Root Mean Square Residual (SRMR) = .034.

Standardized Factor Loadings

Within the optimal three-factor solution, all subtasks loaded significantly onto their designated latent constructs ($p < .001$):

• Factor 1: Core Literacy / Decoding Fluency
  • Timed Word Identification Accuracy: $lambda = .84$
  • Timed Word Reaction Speed: $lambda = -.78$
  • Pseudoword Decoding Speed and Accuracy: $lambda = .89$
  • Orthographic Choice Latency/Accuracy: $lambda = .81$
  • Visual Spelling Verification: $lambda = .76$
• Factor 2: Lexical / Vocabulary Knowledge
  • Receptive Vocational Vocabulary: $lambda = .82$
  • Semantic Synonym Identification: $lambda = .77$
  • Contextual Word Definition Matching: $lambda = .73$
• Factor 3: Arithmetic / Mathematical Fluency
  • Basic Number Fact Retrieval: $lambda = .85$
  • Mental Operations Speed: $lambda = .82$
  • Applied Vocational Math Problems: $lambda = .71$

Inter-factor correlations confirmed that Core Literacy was moderately correlated with Lexical Knowledge ($r = .48$) and Mathematics ($r = .36$), while Lexical Knowledge and Mathematics exhibited a correlation of $r = .41$, validating both their conceptual interrelatedness and structural distinctiveness.

10. Instrument / Measurement Tool

The IDAA-MBO is administered as a structured, interactive, computer-based assessment. Below are the technical and operational specifications of the instrument:

• Target Population: Students aged 15 to 25+ years attending secondary vocational education (MBO levels 1, 2, 3, and 4) in the Netherlands and Flanders.
• Administration Modality: Standard desktop computer or laptop equipped with calibrated visual display, standardized keyboard/mouse input, and high-quality stereo headphones for audio instructions and speech stimuli. Group administration is permissible in controlled, proctored computer labs.
• Total Administration Duration: Approximately 45 to 60 minutes across all modules. Built-in, self-paced resting pauses are positioned between subtests.
• Subtest Architecture and Task Inventory:
  • Module A: Core Reading and Spelling (Dyslexia Battery)
    • Task 1: Timed Word Reading / Lexical Decision (evaluating reaction time and accuracy).
    • Task 2: Pseudoword Decoding (evaluating grapheme-to-phoneme conversion efficiency).
    • Task 3: Orthographic Choice (distinguishing correctly spelled words from homophonic or visually deceptive foils).
    • Task 4: Audiovisual Spelling (typing or identifying spelling variants following spoken word delivery).
  • Module B: Vocabulary and Language
    • Task 5: Receptive Vocabulary (matching target technical/general terms to correct semantic definitions).
    • Task 6: Contextual Word Association (identifying semantic links within vocational narrative prompts).
  • Module C: Arithmetic / Mathematical Competence
    • Task 7: Automated Arithmetic Facts (rapid-fire timed single- and double-digit basic operations).
    • Task 8: Applied Math Problems (vocational multi-step numerical calculation).
• Response Capture and Timing: Millisecond-accurate reaction time logging triggered by keyboard keypresses or precision mouse clicks, combined with automated discrete accuracy tracking (correct = 1, incorrect = 0).
• Scoring and Report Generation: Automated algorithmic generation of raw scores, standard scores ($z$-scores, $T$-scores), and percentile ranks referenced to representative MBO normative cohorts. A composite “Dyslexia Risk Index” is generated alongside a multidimensional profile report highlighting strengths and weaknesses across the reading, vocabulary, and math axes.

11. Permissions & Fee and Test Year

• Year of Publication: 2010 (initial release and national standardization for MBO; revised norm tables released periodically).
• Authors and Copyright Holders: Judith I. Bekebrede, Aryan van der Leij, and the University of Amsterdam / University of Antwerp research consortium.
• Commercial Publisher / Distributor: Published and distributed commercially by Boom Psychologie / Boom Test Uitgevers (Amsterdam, The Netherlands).
• Licensing and Fees: The IDAA-MBO is a proprietary, copyrighted instrument. Access requires institutional licenses or digital test credit vouchers purchased through Boom Test Uitgevers. Per-student administration fees apply.
• User Qualification Requirements: Qualified access is restricted to certified educational psychologists, school counselors, remedial educationalists (orthopedagogen), speech-language pathologists, or institutionally trained psychodiagnostic coordinators conforming to European / Dutch Level B test-user standards.
• Research Use: Academic researchers seeking to administer the IDAA-MBO for non-commercial scientific investigations must obtain formal permissions and licensing agreements from the copyright holder/publisher.

12. References

• Bekebrede, J. I., van der Leij, A., & Share, D. L. (2009). Dutch dyslexic children are impaired in local text comprehension, not in global text comprehension. Scientific Studies of Reading, 13(5), 391–412. https://doi.org/10.1080/10888430903162892
• Bekebrede, J. I., van der Leij, A., Snellings, D. L., & Henneman, K. (2010). Interactieve Dyslexietest Amsterdam-Antwerpen voor het Middelbaar Beroepsonderwijs (IDAA-mbo): Handleiding en Verantwoording [Manual and Psychometric Justification]. Boom Test Uitgevers.
• Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001). DRC: A dual route cascaded model of visual word recognition and reading aloud. Psychological Review, 108(1), 204–256. https://doi.org/10.1037/0033-295X.108.1.204
• Nicolson, R. I., & Fawcett, A. J. (1990). Automaticity: A new framework for dyslexia research? Cognition, 35(2), 159–182. https://doi.org/10.1016/0010-0277(90)90040-Z
• Pennington, B. F. (2006). From single to multiple deficit models of developmental disorders. Cognition, 101(2), 385–413. https://doi.org/10.1016/j.cognition.2006.04.008
• Stanovich, K. E. (1988). Explaining the differences between the dyslexic and the garden-variety poor reader: The phonological-core variable-difference model. Journal of Learning Disabilities, 21(10), 590–604. https://doi.org/10.1177/002221948802101003
• Willcutt, E. G., Petrill, S. A., Wu, S., Boada, R., DeFries, J. C., Olson, R. K., & Pennington, B. F. (2013). Comorbidity between reading disability and math disability: Concurrent psychopathology, functional impairment, and etiology. Journal of Learning Disabilities, 46(6), 500–516. https://doi.org/10.1037/a0026884
• Wolf, M., & Bowers, P. G. (1999). The double-deficit hypothesis for the dyslexias. Journal of Educational Psychology, 91(3), 415–438. https://doi.org/10.1037/0022-0663.91.3.415

13. Items of the Scale