Challenges in Testing for Dyslexia

There are many challenges in the assessment of dyslexia, including the use of varied terminology; reliance on phonological awareness as the sole linguistic risk factor; the inappropriate use of current assessment models; the underdiagnosis of twice-exceptional students; difficulties with early identification; complications with English language learners and the existence of co-occurring disorders that confound an accurate diagnosis. This blog discusses the challenges in testing for dyslexia in-depth, as well as the development of an innovative dyslexia test that can help address some of these challenges.

Challenges in Testing for Dyslexia Highlights:

1. Dyslexia is a clear, diagnosable condition
2. There are challenges in testing for dyslexia, including:
   1. the use of varied terminology in dyslexia testing
   2. reliance on phonological awareness as the sole linguistic risk factor
   3. the inappropriate use of current assessment models
   4. the underdiagnosis of twice-exceptional students
   5. difficulties with early identification
   6. complications with English language learners and the existence of co-occurring disorders that confound an accurate diagnosis
3. Socioeconomic status, trauma, and the disruption in reading instruction caused by COVID-19
4. Tools with good psychometric properties based on recent dyslexia research improve dyslexia assessment
5. The Tests of Dyslexia (TOD) which will be available in 2023

For students with reading problems, one of the most immediate and important goals for educational professionals and parents is to obtain an accurate diagnosis, which should then inform appropriate interventions. Although dyslexia manifests differently in individuals and exists on a continuum from mild to severe, it can be relatively easy to diagnose when a definitive constellation of symptoms (poor word reading, slow reading rate, poor spelling) and a family history are present. Shaywitz (2003) stated: “The diagnosis of dyslexia is as precise and scientifically informed as almost any diagnosis in medicine” (p. 165). Interestingly, over a century ago, Hinshelwood (1917), a Scottish ophthalmologist, came to a similar conclusion, noting that little difficulty exists in the diagnosis of congenital word-blindness (what we now know of as dyslexia) since the general picture of the “condition” is as clear-cut and distinct as any diagnosis in medicine (p. 88).

Despite awareness of this reading disorder and its associated characteristics, examiners often encounter additional challenges during assessment and diagnosis. The purpose of this blog is to discuss several factors that can complicate an evaluation—and at times invalidate the conclusions—as well as suggest ways to address or resolve these challenges. Some major challenges include: (a) use of varied terminology; (b) sole reliance on phonological awareness as the only linguistic risk factor of dyslexia; (c) inappropriate use of current assessment models to diagnose dyslexia; (d) underdiagnosis of twice-exceptional students; (e) difficulties with early identification; (f) different orthographies that complicate the evaluation of English Language Learners; (g) comorbidities that confound accurate diagnosis; (h) additional factors, such as educational opportunity, socioeconomic status, and the disruption in reading instruction caused by COVID-19; and (i) need for tools with good psychometric properties based on recent dyslexia research (Andresen & Monsrud, 2021).

Use of Varied Terminology

Despite the issued memorandum by the United States Department of Education (October 23, 2015) noting that local and state education agencies use the terms dyslexia, dyscalculia, or dysgraphia in describing and addressing unique needs through evaluation, eligibility, and IEP documents, some examiners are reluctant to use the term dyslexia in explaining a student’s reading and spelling difficulties. Siegel and Mazabel (2013) opined: “We do not understand why the term ‘dyslexia’ is often viewed as if it were a four-letter word, not to be uttered in polite company” (p. 187).

Rather than using the term dyslexia, some examiners prefer to use other terms such as a specific reading disability, a learning disability in basic reading skills or reading fluency/rate, a specific reading disorder (2022 ICD-10-CM Code F81.0), or a specific learning disorder with an impairment in reading (DSM-5 315.00). Any of which are also applicable to a student with dyslexia. Although the term learning disability is a broader category and includes additional disorders, such as oral language, mathematics, or writing, many school districts use this term for a student with dyslexia (e.g., a learning disability in basic reading skills or a learning disability in reading rate). If examiners prefer or if their profession requires an alternate term, it is important to incorporate the phrase “also referred to as dyslexia” to reduce confusion among both educational professionals and parents.

Sole Reliance on Phonological Awareness as the Major Linguistic Risk Factor of Dyslexia

Some U.S. state and/or district guidelines identify poor phonological awareness as the only linguistic risk factor for dyslexia. Also, this is true of major organizations that advocate for individuals with dyslexia. For example, the International Dyslexia Association defines dyslexia as “…a specific learning disability that is neurobiological in origin. It is characterized by difficulties with accurate and/or fluent word recognition, poor spelling, and decoding abilities. These difficulties typically result from a deficit in the phonological component of language that is often unexpected in relation to other cognitive abilities and the provision of effective classroom instruction” (Lyon et al., 2003).

Although weaknesses in phonological awareness can contribute to reading and spelling difficulties, a growing body of literature supports a model of dyslexia that includes multiple linguistic risk factors (Compton, 2020). Three core linguistic risk factors for dyslexia are poor phonological awareness, poor verbal memory, and slow rapid automatized naming (RAN; Rose, 2009).

Another linguistic risk factor is poor orthographic knowledge (Mather & Jaffe, 2021). Weaknesses in orthographic knowledge are attributed to difficulty recalling word patterns, limited literacy experiences, and reduced exposure to print. To differentiate between the two, an examiner must ascertain that students in a school setting have had sufficient exposure to print. Apel (2011) specified that orthographic knowledge consists of two levels: lexical, which requires access to stored representations of known words; and sublexical, which requires knowledge of permissible letter patterns and sequences. The results from a recent meta-analysis found that in students with dyslexia, an orthographic knowledge deficit was as large as the deficits in both phonological awareness and RAN (Georgiou et al., 2021). Pennington et al. (2012) noted that if examiners adhere to a single deficit profile, such as using only poor phonological awareness as the sole criterion for determining dyslexia, they will miss about one-half of the cases. Thus, a multifactorial model perspective is critical for accuracy in the identification of dyslexia (Compton, 2020).

Inappropriate Use of Current Assessment Models to Diagnose Dyslexia

Although the results from intelligence tests can be useful in determining a pattern of strengths and weaknesses, these results can also hinder accurate identification. Nearly a century ago, Orton (1925) noted that it seems probable that “…psychometric tests as ordinarily employed give an entirely erroneous and unfair estimate of the intellectual capacity of these children” (p. 582). Because children with dyslexia often earn low scores on specific subtests that are included in intelligence tests, such as measures of processing speed, working memory, and vocabulary, the overall or composite IQ score can be reduced significantly.

This situation is particularly problematic for school districts that rely solely on ability–achievement discrepancies for a learning disability determination; the Full-Scale IQ score for a student with dyslexia will almost invariably be lowered because of their poor performance on subtests measuring specific constructs that are affected by dyslexia. Consequently, examiners should conduct a more comprehensive evaluation and not rely solely on the use of a discrepancy criterion for the identification of dyslexia. Examiners should analyze performance on related measures to obtain data that inform a pattern of performance consistent with dyslexia. They should also analyze factors that are less likely to be impacted by dyslexia (e.g., measures of reasoning, linguistic comprehension, quantitative thinking). Hopefully, most evaluators received training in their coursework regarding accurate test interpretation.

A further problem is that some state and district guidelines assert that a student must have average or above intelligence to have dyslexia. Given that dyslexia is a neurobiological disorder, an individual of any level of intelligence can have dyslexia (Rose, 2009). Furthermore, for older students with dyslexia, their scores on measures of vocabulary and knowledge often decline because of the limited amount of time they have spent reading, the main way we acquire new vocabulary. Three decades ago, Bateman (1992) observed: “The problems in using a formula to identify students who have learning disabilities are many, serious, and too often disregarded” (p. 32).

In addition, other models conceptualized to screen students for reading problems may miss students with dyslexia. For example, Response to Intervention (RtI) models, currently used by many school systems across the country, require that some criteria reflecting poor performance be obtained for students to be considered at risk and in need of multi-tiered intervention. Because some students with dyslexia have well-developed cognitive abilities, other than those affected by dyslexia, they may be able to compensate for their reading-related difficulties and not score low enough on an RtI screening instrument to meet an at-risk criterion. This is particularly the case for twice-exceptional students who are gifted but also have dyslexia.

Underdiagnosis of Twice-Exceptional Students

Twice-exceptional students can have reading scores in the average or better range and still have dyslexia. Because their scores are average, they are not detected by early screenings and do not receive timely interventions. To ensure these students are not overlooked, examiners must consider the student’s level of intelligence, educational history and opportunities, and daily functionality. Some students can compensate for their reading problems, and their reading limitations are unidentified early on. The National Joint Committee on Learning Disabilities advised that “…Although twice-exceptional individuals may appear to be functioning adequately in the classroom, their performance may be far below what they are capable of, given their intellectual ability” (p. 238). In addition to considering a student’s intellectual ability, an examiner should consider high aptitudes in other areas, such as science and mathematics. In her classic book Children Who Cannot Read, Monroe (1932) explained: “The children of superior mental capacity who fail to learn to read are, of course, spectacular examples of specific reading difficulty since they have such obvious abilities in other fields” (p. 23).

In addition to assessing word reading accuracy, examiners need to survey reading rates on grade-level text. Many twice-exceptional students still have compromised reading rates despite average or even above average reading comprehension. As Shaywitz and Shaywitz (2020) explained: “There is no one single test score that ensures a diagnosis of dyslexia. It is the overall picture that matters. An extremely bright child who has a reading score in the average range but who struggles and cannot learn to read fluently […] has dyslexia” (p. 166).

Difficulties with Early Identification

Within the United States, there is no unified system for identifying children in the early years for risk of reading failure. Many districts provide universal screening for reading problems three times a year, but the type of reading specialist support offered in classrooms varies from school to school. Special education teachers are not required to demonstrate proficiency or knowledge of reading instruction, so the quality and intensity of that instruction will vary from school to school. Although a diagnosis of dyslexia becomes more crucial in later years when the student is likely to require accommodations on high-stakes exams, the earlier the identification of this disability, the better. 

The two most prevalent models for early identification in the United States are an ability–achievement discrepancy and a Response to Intervention (RtI) model. Some districts also use a pattern of strengths and weaknesses for early identification, referred to as a PSW approach. The ability–achievement discrepancy model, traditionally used to diagnose a learning disability, is problematic for the early identification of children with dyslexia, as is an RtI model (McCallum et al., 2013; McClurg et al., 2020). Although many school districts offer early screening in kindergarten and first grade, some students with dyslexia have yet to fall behind and do not demonstrate a discrepancy or fail an early screening. The use of a discrepancy model for the early identification of dyslexia results in what Ozernov‐Palchik and Gaab (2016) described as the “dyslexia paradox.” Dyslexia is typically not identified until second grade when the child has not learned to read as expected; the paradox is that early intervention is most effective in Pre-K–Grade 1 prior to reading failure.

Consider the case of Rai, a first-grade student. His scores in reading and spelling fall in the low average to average range. He reverses nearly half of the letters of the alphabet. He had speech-language therapy beginning at age 3½ and continuing until age 5. He has been writing his name since he was 3, yet he still writes a backward R. His father is a cardiologist, and his mother is a psychiatrist. His father describes himself as having similar symptoms when he was young, noting that he “flunked” handwriting and that school “was torture” until he got to college. Although Rai’s father was in the gifted program, his teachers told his parents he was an underachiever. Rai’s older brother, age 10, is in the gifted program. Rai is in an enriched environment with lots of books and family time spent reading. He has dyslexia but would not qualify for services in a district that has a strict reliance on a discrepancy formula or an RtI model because he is not yet far enough behind.  Fortunately, even though he was deemed ineligible for special education, his classroom teachers are aware of his difficulties and are supporting him with a small group reading intervention for 45 minutes each day.

For the accurate early identification of dyslexia, examiners must consider additional factors besides a discrepancy or the results from RtI screening scores. Some young students with dyslexia will not meet an at-risk criterion, either with a discrepancy model or early screening, even though they fall well below the expected performance for gifted or high-ability students. Examiners must also consider whether a student has a history of early speech and language skills/difficulties, a family history of dyslexia, and the level of difficulty mastering early reading and writing skills.

Many school districts have not embraced the use of the term dyslexia with young children. One reason is that it was originally believed to be a medical diagnosis rather than an educational diagnosis, as with ADHD. Another reason is that dyslexia is considered the most common learning disability. The important point is not what diagnostic label to use, but to establish some consistency in terminology and ensure that students who are struggling with reading for whatever reason, receive the support they need.

Different Orthographies That Complicate the Evaluation of English Language Learners

Languages differ regarding their orthography, the writing system. Some languages have deep orthographies with more complex relationships between the speech sounds (phonemes) and the letters that represent these sounds (graphemes) (e.g., English). Other languages have shallow orthographies with higher regularity between the phonemes and graphemes (e.g., Spanish, Finnish).

In languages with deep orthographies, like English, early challenges for students involve accuracy in the development of decoding and encoding (spelling). In languages with shallow orthographies, like Spanish, the main challenge is associated with the development of speed and automaticity with word recognition. For young children learning to read in English, the best indicators of dyslexia are poor phonological awareness, phonics, and spelling. For young readers learning to read in Spanish, phonological awareness, orthographic coding, and RAN are the best predictive measurements (Clinton et al., 2013).

At the start of literacy instruction, these three variables appear to be good predictors of reading in most languages: phoneme awareness, letter knowledge, and RAN. However, in the development of reading ability for languages with shallow orthographies, RAN seems to be the strongest predictor. Thus, the components of a comprehensive evaluation may differ depending on the age and first language of the student.

Comorbidities That Confound Accurate Diagnosis

High comorbidity (two or more disorders in the same person) exists between dyslexia and other learning disorders. About 40% of children with dyslexia will have another learning disorder (Moll et al., 2020). The most common comorbidities include Attention-Deficit/Hyperactivity Disorder (ADHD), dysgraphia, dyscalculia, and language impairments. With young children, it is often difficult to discern if the problem developing reading skills can be attributed primarily to ADHD, dyslexia, or both. Consider Ethan, a first-grade student. Ethan is struggling to learn letters and their sounds. He has been taking Ritalin since age 3 for ADHD. During an evaluation for dyslexia, he had difficulty looking at letters. He would glance at a letter when directed but then quickly shift his attention and comment about an object in the office. When attempting a processing-speed task that required him to circle the matching numbers in a row, his eyes would shift to the ceiling. Ethan will have difficulty learning to read, but until his attention challenges resolve, it is impossible to determine whether he also has dyslexia.

Educational Opportunity, Socioeconomic Status, Trauma, and the Disruption in Reading Instruction Caused by COVID-19

Students can be behind in reading for many reasons, with only one of the reasons being dyslexia. Vast variability exists in reading skills right from the beginning of school. Some children come to school and are already reading. Their parents have shown them how letters represent sounds and taught them to blend those sounds to pronounce words. Some children come to school never having read in their home prior to beginning instruction. Other children are homeless or have suffered trauma or abuse and are not receptive to instruction. In addition, the recent COVID-19 pandemic kept many children out of school for an entire school year. Results from a recent study indicated that the effects on the growth in oral reading fluency of students in Grades 2 and 3 were profound in the 2020 school year (Domingue et al., 2021). Not surprisingly, students in lower-achieving school districts developed reading skills at a slower rate than those in higher-achieving ones.

Need for Tools with Good Psychometric Properties Based on Recent Dyslexia Research

The field needs measures with good psychometric properties that are based on recent dyslexia research (Andresen & Monsrud, 2021) and have the same normative basis (Miciak & Fletcher, 2019). Presently, a comprehensive assessment for dyslexia requires using several different assessments. These assessment batteries have different norm samples, publication dates, and age/grade ranges. They are either standardized tests or rating scales, but not both.

To help fill this need, the authors of the Tests of Dyslexia (TOD; Mather, McCallum, Bell, & Wendling, in press) have developed a comprehensive battery to determine whether an examinee exhibits the major characteristics of dyslexia. The goal was to create one battery of co-normed tests measuring the constructs and factors that should be included in a dyslexia evaluation. The TOD was designed to meet both screening and comprehensive evaluation goals.

The TOD includes measures of the primary areas affected by dyslexia (e.g., word reading, rate, spelling), the major linguistic risk factors (e.g., phonological processing, RAN, working memory), and two cognitive ability measures (vocabulary and reasoning). The TOD also includes co-normed self, teacher, and parent rating scales to help determine current functioning levels and relevant family history. It also contains a guide for instructional recommendations that may be used or copied and pasted into educational supports. The main purpose of the TOD is to provide one integrated system that incorporates test results, rating scale results, and educational interventions.

What happens after Dyslexia Testing?

When a student has been diagnosed with dyslexia, early intervention is critical. Both parents and teachers should ensure reading interventions are intensive, systematic, and provide periodic evaluations of progress; that reading interventions are provided by a trained reading specialist, not a classroom aide or peer tutor; and recognize that interventions will change based on the child’s reading development: specific instruction in phonemic awareness and phonics, structural analysis, spelling, or methods to increase reading fluency.

Unfortunately, in the United States, many teachers still lack training in structured reading methods designed to help students with dyslexia. Or in some instances, they have the training but do not have enough time. Efforts are underway to mitigate this issue and prepare more teachers. For example, the International Dyslexia Accreditation (IDA) program approves universities where the course work aligns with the IDA’s Knowledge and Practice Standards for Teachers of Reading. The International Multisensory Structured Language Education Council (IMSLEC) provides a pathway for independent teacher accreditation and preparation to teach students with dyslexia.

Cruickshank (1977) stated: “Diagnosis must take second place to instruction, and must be made a tool of instruction, not an end in itself” (p. 194). Any evaluation for a student with reading difficulties should address the reasons for those difficulties and provide appropriate interventions.

Conclusion 

As with any disorder, when testing for dyslexia an accurate, early diagnosis is beneficial. Not only for intervention but also for the preservation of self-esteem. Ozernov‐Palchik and Gaab (2016) noted that “This ‘dyslexia paradox’ is detrimental to the well-being of children and their families who experience the psychosocial implications of dyslexia for years prior to diagnosis” (p. 157). In his book My Dyslexia, Schultz (2011) explained: “[The] ignorance of my dyslexia only intensified my sense of isolation and hopelessness. Ignorance is perhaps the most painful aspect of a learning disability” (p. 64). Over eight decades earlier, Dolch (1937) pointed out that “Failure to learn to read as others do is a major catastrophe in a child’s life” (p. 1). A major purpose of the development of the TOD was to support early diagnosis and intervention and thus reduce the emotional toll, impact on self-esteem, and social consequences experienced by individuals with this reading disorder. We need to ensure that children with dyslexia are identified at an early age, understand why reading is so hard for them, provide them with appropriate intensive interventions, and treat them with compassion and understanding throughout their school years.

Author contact: n.mather@att.net

References 

Andresen, A., & Monsrud, M. B. (2021). Assessment of dyslexia—Why, when, and with what? Scandinavian Journal of Educational Research. https://doi.org/10.1080/00313831.2021.1958373

Apel, K. (2011). What is orthographic knowledge? Language, Speech, and Hearing Services in Schools, 42(4), 592–603. https://doi.org/10.1044/0161-1461(2011/10-0085)

Bateman, B. (1992). Learning disabilities: The changing landscape. Journal of Learning Disabilities, 25, 29–36.

Clinton, A., Christo, C., & Shriberg, D. (2013). Learning to read in Spanish: Contributions of phonological awareness, orthographic coding, and rapid naming, International Journal of School & Educational Psychology, 1(1), 36-46. http://dx.doi.org/10.1080/21683603.2013.780195

Compton, D. L. (2020). Focusing our view of dyslexia through a multifactorial lens: A commentary. Learning Disability Quarterly, 44(3), 1–6. https://doi.org/10.1177/0731948720939009

Cruickshank, W. M. (1977). Least-restrictive placement: Administrative wishful thinking. Journal of Learning Disabilities, 10, 193-194.

Dolch, E. W. (1939). A manual for remedial reading. Garrard Press.

Domingue, B. W., Hough, H. J., Lang, D., & Yeatman, J. (March 2021). Changing patterns in the growth in oral reading fluency during the COVID-19 pandemic. Policy Analysis for California Education. Working paper.

Georgiou, G. K., Martinez, D., Vieira, A. P. A., & Guo, K. (2021). Is orthographic knowledge a strength or a weakness in individuals with dyslexia? Evidence from a meta-analysis. Annals of Dyslexia. Published online: March 12. https://doi.org/10.1007/s11881-021-00220-6

Hinshelwood, J. (1917). Congenital word-blindness. H. K. Lewis.

Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). A definition of dyslexia. Annals of Dyslexia, 53, 1–14. https://doi.org/10.1007/s11881-003-0001-9

Mather, N., & Jaffe, L. (2021, September). Orthographic knowledge is essential for reading and spelling. Reading League Journal, 15–25.

Mather, N., McCallum, R. S., Bell, S. M., & Wendling, B. J. (in press). Tests of Dyslexia. Western Psychological Services.

McCallum, R. S., Bell, S. M., Coles, J., Miller, K. C., Hopkins, M., & Hilton-Prillhart, A. (2013). A model for screening twice-exceptional students (gifted with learning disabilities) within a response to intervention (RTI) Model. Gifted Child Quarterly, 57(4), 209–222. https://doi.org/10.1177/0016986213500070

McClurg, V. M., Hassett, N. R., Bell, S. M., McCallum, R. S., & Coladata, B. M. (2020, February). A comparison of two models for dyslexia screening [Paper presentation]. National Association of School Psychologists Annual Convention, Baltimore, MD, United States.

Miciak, J., & Fletcher, J. (2019). The identification of reading disabilities. In D. Kilpatrick, R. M. Joshi, & R. K. Wagner (Eds), Reading development and difficulties: Bridging the gap between research and practice (pp. 159–177). Springer.

Moll, K., Snowling, M. J., & Hulme, C. (2020). Introduction to the special issue “comorbidities between reading disorders and other developmental disorders.” Scientific Studies of Reading, 24(1), 1–6. https://doi.org/10.1080/10888438.2019.1702045

Monroe, M. (1932). Children who cannot read. University of Chicago Press.

National Joint Committee on Learning Disabilities (2011, March). Learning disabilities: Implications for policy regarding research and practice: A report by the National Joint Committee on Learning Disabilities. Learning Disability Quarterly, 34, 237–241.

Orton, S. T. (1925). “Word-blindness” in school children. Archives of Neurology and Psychiatry, 14, 581- 615. https://doi.org/10.1001/archneurpsyc.1925.02200170002001

Ozernov‐Palchik, O., & Gaab, N. (2016). Tackling the ‘dyslexia paradox’: Reading brain and behavior for early markers of developmental dyslexia. WIREs Cognitive Science, 7, 156–176. https://doi.org/10.1002/wcs.1383

Pennington, B. F., Santerre-Lemmon, L., Rosenberg, J., MacDonald, B., Boada, R., Friend, A., Leopold, D. R., Samuelsson, S., Byrne, B., Willcutt, E. G., & Olson, R. K. (2012). Individual prediction of dyslexia by single versus multiple deficit models. Journal of Abnormal Psychology, 121(1), 212–224. http://doi.org/10.1037/a0025823

Rose, J. (June 2009). Identifying and teaching children and young people with dyslexia and literacy difficulties. An independent report from Sir Jim Rose to the Secretary of State for Children, Schools, and Families. DCFS Publications. http://www.education.gov.uk/publications/

Schultz, P. (2011). My dyslexia. W. W. Norton & Company.

Shaywitz, S. (2003). Overcoming dyslexia: A new and complete science-based program for overcoming reading problems at any level. Alfred A. Knopf.

Shaywitz, S., & Shaywitz, J. (2020). Overcoming dyslexia (2nd ed.). Alfred A. Knopf.

Siegel, L. S., & Mazabel, S. (2013). Basic cognitive processes and reading disabilities. In H. L. Swanson, K. R. Harris, & S. Graham (Eds.), Handbook of learning disabilities (2nd ed.) (pp. 186–213). Guilford Press.

United States Department of Education, Office of Special Education and Rehabilitative Services. (October 23, 2015). Dear colleague: Guidance on dyslexia.

Acknowledgments 

I would like to thank my co-authors on the TOD (R. Steve McCallum, Sherry M. Bell, and Barbara J. Wendling) for their feedback and suggestions on the initial draft of this blog, as well as Laura Wallof, Stephanie Roberts, Kristen Porter, and Bonnie Mills from WPS, Julia Kender from PATOSS, and Irene Gonzalez.

Related:

Dyslexia and reading comprehension

Video: 10 Minutes: Dyslexia Screening by Your Classroom Teacher Partner

(DESD) Decoding-Encoding Screener for Dyslexia