Reading and Spelling Difficulties and Self-Confidence in Adolescents with an Early or Late diagnosis of Dyslexia
Masterthesis Orthopedagogiek, Pedagogische en Onderwijskundige Wetenschappen, Universiteit van Amsterdam (UvA, University of Amsterdam) Name student: S. E. (Sanne) Blessing Supervisor: mw. dr. M. (Madelon) van den Boer MSc Second supervisor: dr. E. H. (Elise) de Bree Amsterdam, (April, 2017)
Preface
Before you lies the thesis ‘Reading and Spelling Difficulties and Self-Confidence in
Adolescents with an Early or Late Diagnosis of Dyslexia.’ The thesis is written to fulfill my graduation requirements of the study Orthopedagogiek at the University of Amsterdam (UvA). I was engaged in researching and writing this thesis from February 2016 to April 2017.
The project were my data for my thesis is collected was the project Grip op Dyslexie (Grip on Dyslexia) from Loes Bazen. My research questions were formulated by myself and discussed with my supervisor, Madelon van den Boer. Overall the research and writing have taken an extended time, but the investigation has allowed me to answer the question that was
identified. I had contact with Madelon van den Boer after every part I had written, she gave me helpful feedback to improve my writings. Loes Bazen lead the project and guided me to set up the first part of the project, she gathered many participants together with Selma Adriaanse.
I would like to thank my supervisors for their excellent guidance and support during this process. Also many thanks for all the participants, without whose cooperation I would not have been able to conduct this analysis.
Further I would like a particular note of thanks for my friends and parents: who gave me motivation, checked my thesis and helped me with ideas.
I hope you enjoy reading. Sanne Blessing
Contents
Preface... 2
Abstract ... 4
Samenvatting... 5
Reading and Spelling Difficulties and Self-Confidence in Adolescents with an Early or Late diagnosis of Dyslexia ... 6 Method ... 9 Participants ... 9 Measures... 10 Word-reading fluency ... 10 Pseudoword-reading fluency ... 11 Spelling ... 11 Reading comprehension ... 11 Self-confidence ... 12 Procedure ... 12 Data analysis ... 13 Results ... 13
Differences among the Early, Late and Control groups ... 15
Reading ability ... 15 Spelling ... 16 Reading comprehension ... 16 Self-confidence ... 16 Discussion ... 17 References ... 24
Abstract
Among the increasing number of high school students with dyslexia, the distinction can be made between those with an early diagnosis and those with a late diagnosis. To obtain a better perspective of the performance of these two groups, the current study illustrates the difference in their abilities in reading, spelling, reading comprehension, and in their sense of self-confidence. Participants were 53 10th-grade students from eight Dutch high schools. The performance on word reading fluency, pseudoword reading fluency, spelling, reading comprehension, and self-confidence was compared among 20 students with an early diagnosis, 15 with a late diagnosis, and 18 normally achieving students. The results demonstrated that the early diagnosed group had both reading and spelling difficulties, whereas the late diagnosed group had only reading difficulties; no significant differences were found among the groups in term of reading comprehension and self-confidence. The only difference between the groups with dyslexia is found for spelling. Despite reading difficulties in both dyslexic groups, and also spelling difficulties for the early diagnosed group, none of the students had low self-confidence or problems understanding the meaning of a text. Extra help in school for 10th-grade students with dyslexia remains important with regard to their reading abilities, and teachers should maintain an awareness of students’ spelling, reading comprehension, and self-confidence.
Keywords
Samenvatting
Binnen de toenemende groep studenten met dyslexie kan een onderscheid gemaakt worden tussen een vroege diagnose en een late diagnose van dyslexie. Om een beter beeld te krijgen van de prestaties van deze twee groepen, toont de huidige studie de verschillen tussen hun vaardigheden in lezen, spellen, begrijpend lezen en gevoel van zelfvertrouwen. De participanten waren 53 leerlingen uit de vierde klas van acht verschillende Nederlandse middelbare scholen. Het onderzoek heeft de prestaties op lezen van losse woorden, pseudowoord lezen, spelling, begrijpend lezen en zelfvertrouwen vergeleken voor; 20 leerlingen met een vroege dyslexie diagnose, 15 met een late dyslexie diagnose en 18 leerlingen zonder dyslexie. De resultaten toonden aan dat de vroeg gediagnosticeerde problemen hadden met zowel lezen als spellen, terwijl de laat gediagnosticeerde alleen problemen hadden met lezen. Voor begrijpend lezen en zelfvertrouwen werden geen significante resultaten gevonden. Het enige verschil tussen de groepen met dyslexie is gevonden op spellingsprestaties. Ondanks de moeilijkheden met lezen en de vroeg gediagnosticeerde ook met spellen, hadden de studenten geen problemen met zelfvertrouwen en het begrijpen van een tekst. De extra hulp op de middelbare scholen voor dyslectische leerlingen uit de vierde klas blijft belangrijk voor hun leesvaardigheid en leerkrachten zullen aandacht moeten blijven besteden aan de spelling prestaties, het begrijpend lezen en het zelfvertrouwen van alle leerlingen.
Sleutelwoorden:
Reading and Spelling Difficulties and Self-Confidence in Adolescents with an Early or Late diagnosis of Dyslexia
In recent years, there has been an increase in the number of Dutch high school students with a diagnosis of dyslexia. Almost 14% have received such a diagnosis (Sontag & Donker, 2010- 2012), a percentage much higher than previous estimates of, about 3.6% (Blomert, 2005). This rising number of dyslexic students makes it difficult for schools to tent appropriately attention to all their needs. Furthermore, the cost of diagnosing and of devising adequate treatment is an expensive procedure. Because two groups of dyslexics are found in the high schools, early and late diagnosed, gaining better perspective on whether both types of student need the same kind of attention is important to secure the cost-efficiency of the process. The current study first investigates the development related on reading, spelling and reading comprehension and, second, the self-confidence of 10th-grade students with early- and late- diagnosed dyslexia.
Dyslexia is a specific learning disorder characterized by a persistent problem in accurate or fluent word recognition, poor decoding, and spelling (American Psychiatric Association, 2013). This study will use the definition of the DMS-5, because the focus is on reading and spelling. The definition is partly in line with the Stichting Dyslexie Nederland (SDN; Foundation for Dyslexia of the Netherlands, 2016) that is used for diagnosing dyslexia. For students who are diagnosed early it is possible to receive financial support from the government, but to do so they must comply with a stricter definition of dyslexia than that of the SDN (2016). Instead, the criteria are used from Protocol Dyslexie Diagnose en Behandeling (PDDB; Protocol Dyslexia Diagnosis and Treatment) (Blomert, 2005). This group of students must perform within the lowest 10% or lowest 16% in reading of standard average, if performing in the lowest 16% of reading than the spelling score has to be within the lowest 10% of the standard average. Such low performance arises from persistent problems in phonological processing, which are not expected based on the students’ intelligence and the amount of education they received (Blomert, 2005). In both definitions, difficulties in spelling are not necessary for a diagnosis. However, the current study demonstrates the performance of high school students with a diagnosis of dyslexia based on the American Psychiatric Association (2013) definition, with a focus on reading and spelling.
Difficulty in reading comprehension is not considered a core characteristic of dyslexia, which is primarily concerned with word recognition. Performance in reading and spelling, however, can influence reading comprehension. A simple definition of reading assumes that language comprehension and decoding are necessary to understand the meaning of a text
(Tunmer & Chapman, 2012), and students with dyslexia have difficulties with decoding that can make it more difficult to understand a text. For its part, spelling requires the precise writing of the words, requiring word-specific knowledge and knowledge of phoneme-grapheme connections, spelling rules, or morphological rules (Vaessen, 2010). These requirements for spelling may lead to more complete word representations than those learned from reading, making it easier to understand a text (Retelsdorf & Köller, 2014). Typically developed spellers may have more complete word representations than dyslexic spellers. Based on these findings, the expectation is that students with dyslexia have more difficulties with reading comprehension than students without dyslexia.
Furthermore, reading comprehension is valuable when studying the consequences of dyslexia. Leach, Scarborough, and Rescorla (2003) found that late-identified cases were a heterogeneous group: 35% had word-level processing deficits in combination with adequate comprehension skills, 32% had weak comprehension skills accompanied by adequate lower-level skills, and 32% exhibited both kinds of difficulties. The study selected students based on reading difficulties, that were indicated by school and parents of these students. The selection by Leach et al. (2003) is different in comparison to the current study, where the focus is on an official diagnosis of dyslexia based on the aforementioned definition, in which students can have reading and spelling difficulties. Although this difference the current study views a possibly dyslexic not only those with word-level processing deficits in combination with adequate comprehension skills but also those with both kinds of difficulties. Because the study of Leach et al. (2003) found decoding and reading comprehension difficulties, and reading comprehension is influenced by the performance in reading and spelling, it is included as a variable in our study.
Previous studies also found a distinction between early- and late-identified students. Catts, Tomblin, Compton, and Bridges (2012) researched children from kindergarten to 10th -grade and focused on word reading and reading comprehension by poor readers. The researchers found that late-emerging poor readers show adequate performance in reading and spelling in early school grades; the reason they are late-identified is that they showed difficulties in a later phase. This late development may have to do with the fact that in high schools the expectations of the students’ performance are higher than in primary school; students are expected to have a deeper understanding of a text as well as the challenging concepts they already learned (Leach, et al., 2003). Leach, et al. (2003) examined students with early identified reading disabilities and late identified reading difficulties and their performance on reading comprehension and word reading. For students with a late identified
reading difficulty, there are specific difficulties not only with the decoding and spelling of complex words, but also with the comprehension of written material. Furthermore, the early and late identified groups have about equal deficits in word-level reading abilities, pseudoword decoding, and spelling (Leach et al., 2003). Additionally, Torppa, Eklund, van Bergen, and Lyytinen (2015) studied four groups of Finnish students on the stability of their dyslexia status based on reading fluency from second grade to eighth grade. They found that late-identified students had an unclear cognitive profile at an earlier stage, which made these students more difficult to predict. These previous studies reveal that the difficult predicted late-identified students experience difficulties in a later phase that are similar to the difficulties that early identified students experience. They selected on poor readers, reading difficulties or students with a familiar risk, this selection focused on reading. The current study makes the same distinction, but focuses on official diagnosed dyslexic high school students, where difficulties in reading and spelling will be illustrated. From the findings of the previous studies, the time difference between the early diagnosed and late diagnosed group raises questions; for example, do children with a late diagnosis have the same difficulties as students with an early diagnosis? Or do late-diagnosed experience a smaller range of difficulties?
Although dyslexic students can have difficulties in reading and writing, some suffer more from their difficulties than others do. Elbro’s study (2010) makes a distinction between dyslexia as a disability and dyslexia as a handicap. First Elbro (2010) determined whether someone had the disability of dyslexia by examining their test scores. The disability may influence school performance, because the student is experiencing reading and writing difficulties. Second, Elbro (2010) investigated the handicap, that is, how a person valued the difficulties caused by the disability. These values indicate how much someone suffers from the disability, irrespective of the severity of the difficulties themselves. For example, a student at a lower level of education may suffer less than a student at a higher level, because the expectations within the student’s environment are different. Elbro (2010) found that adults’ perceptions of themselves as dyslexic depend on education and social context more than their initial reading difficulties, as vocabulary skills. Students with a larger vocabulary had a higher risk of becoming more aware of their handicap in school (Elbro, 2010).
Furthermore, dyslexic people’s perception of their abilities in reading and spelling can influence their well-being. This perception can be defined as their self-confidence (Burden, 2008). They may have the feeling that they are different than their peers or other people (Lawrence, 2006). Perry (2011) describes self-confidence as context-specific to particular
tasks. People seem to display self-confidence through a wide range of activities; it is different for everyone and a part of children’s development (Perry, 2011), which make self-confidence hard to define. In the current study, self-confidence is defined as the view of one’s abilities and the trust in one’s performance, whether one has a disability or not. The focus is on one’s own abilities in school subjects such as reading and writing. Gunnel Ingesson (2007) interviewed Finnish students with dyslexia about their experience in school. The researcher found that students’ self-esteem could drop during the period they failed or fell behind in school, before they were diagnosed with dyslexia. Such negative experiences cannot be avoided completely, but with support from parents and professionals, the frequency and intensity can be reduced (McNulty, 2003). Hoy, Gregg, Wisenbaker, Manglitz, King, and Moreland (1997) and Carroll and Iles (2006) found that college students with learning disabilities (LD) reported more symptoms of anxiety than non-LD college students did. Based on previous research, it can be expected that students with dyslexia who experience difficulties in school have lower self-confidence than students without difficulties.
In sum, the number of high school students diagnosed with dyslexia is increasing (Blomert, 2005; Sontag & Donker, 2010-2012). This rising number raises questions, as how can all these student tent appropriately attention? Within high school the group can be divided into an early diagnosed and a late diagnosed group. This study investigates the reading and spelling performance of an early diagnosed and a late diagnosed group and it examined their self-confidence regarding their reading and spelling abilities. The aim is to better understand how to help both groups. Based on our definition of dyslexia and on previous literature it can be expected that students with dyslexia have difficulties with reading, spelling, reading comprehension, and self-confidence whereas students without dyslexia have fewer difficulties. To investigate the differences in performance, three groups were compared: early diagnosed dyslexia, late diagnosed dyslexia and a control group. First the current study analyses the differences among these three groups in terms of reading, spelling, reading comprehension and self-confidence. Second it illustrates the differences between the group with an early diagnosis and a late diagnosis based on these variables.
Method Participants
In this study, 53 Dutch 10th-grade high school students (28 boys) aged between 14.8 and 18.4 years old participated. They were divided into three different groups: early diagnosis of dyslexia (‘Early’, 10 girls, 10 boys, age range: 14.8–18.4 years), late diagnosis of dyslexia (‘Late’, 6 girls , 9 boys, age range: 14.8–17.8 years), and the control group (‘Control’, 9 girls,
9 boys, age range: 15.6–17.3 years). The students within the Early group had received their official diagnosis in primary school (grade 1 to 6), whereas the Late group had received their official diagnosis after primary school (grade 6 to 10) and the Control group had no diagnosis of dyslexia. Within the groups, boys and girls were equally distributed (see Table 1). Also the mean age did not differ across groups (F(2, 50) = 0.981, p = 0.038).
Differences in gender of the students across groups were checked with a chi-square test. These differences were not significant (χ2 = 0.432. df = 2, p = 0.806). The students came from different school levels. Within the Dutch school system, there are three school levels: pre-vocational education (VMBO), general secondary education (HAVO), and preparatory scientific education (VWO). As shown in Table 1, most students were from HAVO/VWO. The small number of VMBO students was because the study started when these students had their final exams.
Table 1
Student Characteristics For Each Group
Age (in years) School level
Group N % Boys M SD % VMBO % HAVO % VWO Early 20 50.0 16;5 9.9 15.0 55.0 30.0 Late 15 60.0 16;6 9.6 6.7 40.0 53.3 Control 18 50.0 16;2 5.9 5.6 38.9 55,6
Note: Vocational education (VMBO), general secondary education (HAVO), and preparatory scientific education (VWO).
Measures
Word-reading fluency: The One-Minute Test was used to measure word decoding (EMT; Brus & Voeten, 1999). The student was asked to read quickly and accurately as many words as possible within one minute. Within dyslexia research this task is used—both in science and in diagnosis—as a measure of decoding difficulty as a characteristic of dyslexia. The card contained 116 words of increasing difficulty; for example, an easy word was ‘doel’ (goal), and a difficult word was ‘promenade’ (promenade). The score consisted of the number of correctly read words. The maximum score was 116. The parallel test-retest reliability for the first grade of high school was .87 (Evers, Egberink, Braak, Frima,Vermeulen, & Van Vliet-Mulder, 2009-2012). The norms for general secondary education (HAVO) of the
One-Minut Test were used to illustrate the number of dropouts in each group (Kuijpers, van der Leij, Been, van Leeuwen, ter Keurs, Schreuder, & van den Bos, 2003).
Pseudoword-reading fluency: The Klepel was used to measure decoding of nonwords (de Klepel; Van den Bos, Lutje Spelberg, Scheepstra, & de Vries, 1994). The student was asked to read nonwords as quickly and accurately as possible within two minutes. The card contained 116 nonwords that ranged from short/easy words like olg to longer/more difficult words like meschauwank. The score consisted of the number of nonwords that were read correctly. The maximum score was 116. Parallel test-retest reliability reported for first grade of high school was .89 (van den Bos et al., 1994). Also the norms for HAVO of this task were used to illustrate the number of dropouts in each group for the Klepel (Kuijpers, van der Leij, Been, van Leeuwen, ter Keurs, Schreuder, & van den Bos, 2003).
Spelling: The wonderlijke weer (wonderfull weather) measured the student’s spelling ability. Dutch sentences were dictated by the test assistant. This dictation was to signalling difficulties in spelling and included sentences which increased in difficulty (KPC groep, 2005). The task was intended to write down the spoken sentences. There were 10 sentences, for example, ‘Lang geleden dacht men/ dat goden/ de sterren en planeten bestuurden’ (For a long time it was thought/ that gods/ ran the stars and the planets). On every ‘/’, there was a pause in the speech of the test assistant, and the student wrote the word(s) down as accurately as possible. The total score comprised all incorrectly written words; the best score was 0, and the maximum number of incorrect words was 178. High scores on spelling meant that the student made more mistakes, whereas a lower score in spelling meant that students made less mistakes. For this task the norms were used to illustrate the dropouts on spelling, but for this task there was no reliability known (KPC groep, 2005).
Reading comprehension: The reading comprehension task consisted of two short texts that were designed for the purpose of this study. The students were asked to read and answered eighteen questions about the texts, which were taken from the final exam in Dutch as administered in high schools. Students identified the type of text, the meaning of the text, relationships among parts of the text, and recognised positions and types of arguments (Examenblad, 2016). For example, the question ‘Hoe wordt het onderwerp in deze paragraaf geïntroduceerd?’ (How is the subject introduced in this paragraph?) was followed by a multiple-choice answer option from A to D. Every correct answer was awarded one point. There were eight multiple-choice questions and 10 open-ended questions. The number of points a student received for the open-ended questions depended on the answer (ranging from 0–4). The maximum score on this task was 27.
Self-confidence: Self-confidence was examined with a questionnaire that was specifically designed for this study. There were nine questions about self-confidence, and the participant chose the answer that fits the best for them. An example of a question is, ‘Ik ben regelmatig bang dat de teksten die ik schrijf niet goed genoeg zijn’ (I can be afraid that a text that I wrote is not good enough). The five answer options were: ‘perfectly applies to me’, ‘applies to me’, ‘applies a little to me’, ‘does not apply to me’, and ‘does not at all apply to me’. These were judged on 5-points scale (1 = does not at all apply to me and 5 = perfectly
applies to me). The maximum score was 45 and the minimum was 9. When a student had a
high score on the questions, the student had low self-confidence. Besides the multiple-choice questions, there were three questions with a different type of score. These questions had a scale from 1 to 100, and the student placed a pointer on the number that suited them best. An example of this sort of question was, ‘How badly does it bother you to read more slowly than your classmates?’ The score of 0 was ‘suits me not at all’, and a score of 100 was ‘suits me very much’. These questions were individually analysed and described in the text. For these three questions the reliability is checked. The three questions had a Cronbach’s alpha of .463. This represented an unacceptable correlation, which meant that these questions had to be interpreted carefully.
Procedure
Within this study eight Dutch high schools were asked to take part in the study. Contact within the school was mostly with the school’s care coordinators, because they had contact with all students with dyslexia. When a school agreed to participate in the study, they contacted the parents of the children with dyslexia. The parents indicated through an active consent form that their son or daughter could participate in the study. Next, the children with dyslexia were asked to confirm that they wanted to participate in the study. If the students also agreed, then the test assistants made an appointment for testing. The research had three sessions in total. All sessions of the study took place in the school, in a quiet place where the children could concentrate on the tasks at hand. The Control group was selected by the group with dyslexia: The students with dyslexia were asked to bring a friend who did not have dyslexia. The parents of the Control group students also agreed through an active consent form that their son or daughter would participate in the study. Next, the research assistant made an appointment with the students in the control group. The consent was obtained by file number 2015-CDE-4749 of the project Grip op Dyslexie (Grip on Dyslexia) by the Ethics Committee.
The current study was part of a larger research project, including many tasks. All the tasks were divided into parts A, B, and C. Subcategories were reading and spelling tasks, cognitive tasks, and executive function tasks. For all 21 tasks of the study, the total duration was three hours. The tasks were divided within three session, every part of the study lasted one hour. Most students did all three parts on a different day. The sessions A and B often took place in rapid succession. Session C was planned later than session A and B. The average time to complete all sessions was five weeks, depending on when session C was planned. This study only included measurements of word reading fluency, pseudoword reading fluency, spelling, reading comprehension, and self-confidence. Word-reading fluency and pseudoword reading fluency were in part A and were administered individually. Reading comprehension and spelling were in part C and took place in a group setting. The self-confidence questions were included in the questionnaire and could be answered online at any time during the study.
Data analysis
Differences were examined in performance on reading and spelling of students with an early diagnosis of dyslexia, children with a late diagnosis, and controls. Before the analyses, the data was checked for missing data, outliers, and assumptions of the analysis of variance (ANOVA). Also, the correlations among all the tasks were examined. To analyse the data, MANOVA was used for word reading fluency and pseudoword reading fluency and an ANOVA for spelling, reading comprehension and self-confidence. The MANOVA and ANOVA results only reflect the overall differences, so the differences in scores among the groups were checked with a post-hoc test (Bonferroni). This study had five dependent variables: word reading fluency, pseudoword reading fluency, spelling, reading comprehension, and self-confidence. The groups could be compared with each other on all five dependent variables.
For the assessment of significance, the alpha level of .05 was used. Besides that, the effect size was reported in a partial Eta squared. The effect size of the partial Eta square ranged from .01 = small effect, .09 = middle effect, and .25 = substantial/large effect (Cohen, 1998).
Results
Before analyses were run, data were checked for missing values and outliers. The data had some missing values; one spelling score together with one reading comprehension score, and nine questionnaires on self-confidence. The spelling and reading comprehension scores were missing because this student missed the last session. After all of the sessions were
finished, the students were asked to complete the online questionnaire, which included the self-confidence questions. Nine of the students forgot to fill in the questionnaire. No outliers were found, as all of the scores were between 3 standard deviations from the mean of each variable. Scores on all the variables were normally distributed. This was checked with the skewness and kurtosis index.
Table 2 Descriptive Statistics Total sample (N = 53) N M (SD) Range dropout (%) Early WRF 20 73.05 (11.27) 50–92 10 (50) PRF 20 57.80 (12.28) 24–74 11 (55) Spelling 20 10.90 (5.28) 2–19 3 (15) RC 20 13.05 (4.71) 4–23 SC 15 23.67 (5.38) 13–32 Late WRF 15 76.27 (10.79) 59–99 5 (33.3) PRF 15 64.67 (6.97) 47–73 5 (33.3) Spelling 15 6.33 (4.01) 1–13 0 RC 15 14.27 (5.82) 2–22 SC 14 22.36 (7.66) 9–36 Control WRF 18 92.06 (14.87) 60–116 1 (5.6) PRF 17 84.88 (20.68) 47–114 1 (5.6) Spelling 17 3.824 (3.68) 0–13 0 RC 17 16.47 (5.69) 5–24 SC 15 21.00 (4.17) 15–28 Total WRF 53 80.42 (14.93) 50–116 16 (30.2) PRF 52 68.63 (18.50) 24–114 17 (32.1) Spelling 52 7.27 (5.33) 0–19 3 (5.7) RC 52 14.52 (5.46) 2–24 SC 44 22.34 (5.83) 9–36
Note: For spelling a higher meant more mistakes. WRF = word reading fluency; PRF = pseudoword reading fluency; Spelling = wonderfull weather (WW); RC = reading comprehension; SC = self-confidence
The descriptive statistics are presented in Table 2. For word reading fluency and pseudoword reading fluency the Control group had higher scores than the Late group and Early group. Between the Late group and Early group there were only small differences in scores. Spelling scores were different than the other scores, because they were based on the number of incorrectly spelled words, so a lower score is less mistakes in spelling. The Control group had a lower score on spelling than the Late group, and the Early group had the highest score. There were small differences among the groups for reading comprehension and self-confidence. All the variables had a sufficient range of scores.
To illustrated dropouts among the groups, the norm scores for word reading fluency, pseudoword reading fluency and spelling were added. The Early group had for reading fluency, pseudoword reading fluency and spelling the highest number of dropouts. Within the Late group were less dropouts on these tasks than the Early group, but still five students achieved within the lowest 10% compared with the standard averaged of word reading fluency and pseudoword reading fluency. The Late group had no dropouts for spelling. The Control group had only one dropout for word reading fluency and pseudoword reading fluency. A possible reason for the dropout in the Control group is that the norms for HAVO were used and there was one student of VMBO in the control group.
Table 3
Correlations Between Dependent Variable and the Groups
Name task 1 2 3 4 5 1. WRF - 2. PRF .781** - 3. Spelling -.432** -.424** - 4. RC .485** .502** -.480** - 5. SC .021 .040 .240 -.204 -
Note. For spelling a higher meant more mistakes. WRF =word reading fluency; PRF = pseudoword reading fluency; Spelling = wonderfull weather (WW); RC = reading comprehension; SC = self-confidence *** = p < .001
The correlations between the variables are presented in Table 3. Most correlations between the tasks were statistically significant (p < .001) and ranged from -.42 to .79. A strong correlation was found between word reading fluency and pseudoword reading fluency (r = .78). Spelling and reading comprehension correlated moderately (r = -.48). Spelling had negative correlations because the score was the number of mistakes that the student had made. Which meant that students with a high score on spelling made more mistakes. The correlations with self-confidence were all non-significant.
Differences among the Early, Late and Control groups
Reading ability: The first research question was ‘Do the Early group and the Late group have the same difficulties in reading, whereas the Control group has less difficulties in reading?’ Significant differences were found for word reading fluency and pseudoword fluency (Wilks Lambda = .570, F(4, 98) = 7.948, p < .001). Post hoc comparisons for word reading fluency showed that the Late group scored significantly lower than the Control group
(p < .001), the Early group scored significantly lower than the Control group (p = .002), but the Late group and Early group did not differ significantly from each other. For pseudoword reading fluency the post hoc comparisons revealed that the Late group scored significantly lower than the Control group (p < .001), the Early group scored significantly lower than the Control group (p = .001), but the Late group and the Early group did not significantly differ from each other (p = .508) .
Spelling: The second research question was, ‘Do the Early group and the Late group both have difficulties with spelling, whereas the Control group has less difficulties in spelling?’ Differences in spelling were significant (F(2, 49) = 12.07, p < .001). Post hoc comparisons showed an interesting difference between the groups: The Late group made significantly fewer mistakes than the Early group (p = .013), the Early group made significantly more mistakes than the Control group (p < .001), but there was no significant difference between the Late group and the Control group. For spelling, the results were different than for reading. The Early group had more difficulties than both the Late group (p = .013) and the Control group (p = .001). Surprisingly, the Late group had almost the same spelling performance as the Control group (p = .354).
Reading comprehension: The third research question was, ‘Is there a difference in reading comprehension between the Early group and the Late group, whereas the Control group has fewer difficulties in reading comprehension?’ There were no significant differences in reading comprehension (F(2,49) = 1.89, p = .163). Post hoc comparisons were not required. For the reading comprehension task, there were no significant differences between the groups; the scores for this task were almost the same among all the groups.
Self-confidence: The last research question was ‘Does the Early group have a lower confidence than the Late group, whereas the Control group has less difficulty with self-confidence?’ For self-confidence, no significant differences were found (F(2, 41) = .774, p = .468). The scale of self-confidence was designed for the purpose of this study, so the reliability is checked. The self-confidence scale had a Cronbach’s alpha of .822. This indicated that the scale of self-confidence was a reliable measure of self-confidence. Without the question, ‘Ik durf werkstukken of teksten regelmatig niet in te leveren omdat ik bang ben dat ze niet goed genoeg zijn’ (‘I don’t have the courage to hand in assignments or texts, because I’m afraid they are not good enough’) the Cronbach’s alpha was even higher
(α=.843). The score of self-confidence was recalculated without this question, but still no
Besides the nine questions about self-confidence that were answered using a Likert-scale, the questionnaire also had questions where the students could indicate their self-confidence on a scale of 0 to 100. The questions were (1) ‘Ik vergelijk mijn schoolresultaten vaak met die van klasgenoten’ (‘I often compare my school results with those of my classmates’), (2) ‘Als ik tegelijk met klasgenoten iets moet lezen, gaat dat bij mij vaak een stuk langzamer’ (‘When I read together with classmates, I read much more slowly’), and (3) ‘Als ik langzamer lees dan mijn klasgenoten, voel ik me vervelend’ (‘When I read more slowly than my classmates, I feel bothered by it’). The interpretation of these questions had to be careful, because the Cronbach’s alfa was unacceptable.
In response to the first question, the Early group (M = 53.88, SE = 5.82), the Late group (M = 50.71, SE = 7.58), and the Control group (M = 61.93, SE = 6.81) had almost the same results. The ANOVA showed no significant differences on this question (F(2,42) = .729,
p = .488).
Regarding the second question, the Early group (M = 73.40, SE = 7.94) had the highest score compared to the Late group (M = 58.50, SE = 8.41) and the Control group (M=13.80, SE=4.11). However, the Late group had a higher score than the Control group. These differences were significant (F(2, 41) = 19.351, p < .001). Post hoc comparisons showed that the Early group scored significantly higher than the Control group (p < .001), the Late group scored significantly higher than the Control group (p < .001), but that Late and Early group did not significantly differ from each other p = .516 As expected, the Early group and the Late group provided a significantly higher score than the Control group on reading more slowly when reading with a classmate.
The last question was only relevant for students with dyslexia and therefore only assessed in the two dyslexia groups. The Early group (M = 45.87, SE = 9.04) had a higher score than the Late group (M=10.93, SE=5.29). This difference was significant (F(1, 27) = 10.723, p = .003). The Early group had a significantly higher score than the Late group on feeling more annoyed about reading more slowly.
Discussion
This study investigated difficulties in reading, spelling, reading comprehension, and self-confidence among 10th-grade high school students with and without dyslexia. The purpose of this study was to obtain a better understanding of the reading and spelling performance of dyslexic children in the 10th-grade and of their self-confidence regarding reading and spelling. It is important to have a better understanding of the performance of students with dyslexia because in recent years the number of high school students with a
diagnosis of dyslexia is increased (Blomert, 2005; Sontag & Donker, 2010 - 2012). As a result of the increase, it is difficult for schools to give the appropriate attention to all these students. By obtaining a better understanding of the effects on their performance the specialist care could be more attuned to the two groups with dyslexia. For the research sample three groups were delineated: students with an early diagnosis of dyslexia, students with a late diagnosis of dyslexia, and students without dyslexia as the control group. The threshold for the early group, was that the individual was diagnosed with dyslexia in primary school (grade 1-6), whereas the late group received their diagnosis anywhere after this moment. The difference with previous studies was that the current study selected on 10th -grade high school students with an official diagnosis of dyslexia, whereas other studies focused on reading disabilities or poor readers.
In the Netherlands students with an early diagnosis of dyslexia can receive financial support from the government for the diagnosing dyslexia and specialist treatment. In order to receive this support, these students have to meet a number of criteria that are stated in the ‘Protocol Dyslexie Diagnose en Behandeling’ (PDDB; Protocol Dyslexia Diagnosis and Treatment). One of these criteria stated that the student had to achieve within the lowest 10% of the standard average in reading or lowest 16% for reading and for spelling in lowest 10% of the standard average. This study illustrated the norm scores for word reading fluency, pseudoword reading fluency and spelling to demonstrate if students achieved within this group. Around 50% of the group with an early diagnosis scored within the lowest 10% on reading and 15% on spelling. Within the late diagnosed group 33.3% scored within the lowest 10% achievers on reading, but no one of this group in the lowest 10% for spelling. The control group had only 5.6% that fell in the lowest 10% of reading. This difference in percentages was expected, since solely the group with an early diagnosis was eligible for financial support from the government, and therefore should comply with these criteria.
The first finding was that the two dyslexic groups obtained lower scores on word reading fluency and pseudoword reading fluency than the control group. No difference was found between the two dyslexic groups; they both had equal difficulties with reading. These findings were expected, as difficulties in reading are part of the diagnosis of dyslexia. People with developmental dyslexia have difficulties with tasks that require word-level decoding, and word and pseudoword reading fluency tasks belong to this category (Hulme & Snowling, 2016). Leach et al. (2003) found that a late-identified group and an early-identified group had equal deficits in many word-level reading abilities, including accuracy of word recognition and pseudoword decoding. The results of this study are thus in line with the findings of Leach
et al. (2003). Although difficulties in reading of students with a late diagnosis of dyslexia are recognized in a later stage, the current study found that the reading performance of the late diagnosis group were almost the same as the group with an early diagnosis. Catts et al. (2007) demonstrated that the difficulties in reading for the late-identified group evolve at a later phase. For the current study, this means that although their difficulties evolved later, the group with a late diagnosis did not have less difficulties in reading than the students with an early diagnosis. In other word, the time of diagnose does not affect the eventual effects on performance in reading.
Regarding spelling, the results were rather different: This study has found a difference in spelling scores between the group with an early diagnosis and the group with a late diagnosis. The early diagnosed group scored lower than the control group, but the group with a late diagnosis had nearly the same scores as the control group. These results are surprising; based on the definition of dyslexia in this study, it would be expected that both groups – both diagnosed with dyslexia – would score nearly the same on spelling. However, with a diagnosis of dyslexia as defined in SDN (2016), a person has a reading and/or spelling deficit, so it is not necessary to have difficulties in spelling to have a diagnosis of dyslexia. Leach et al. (2003) revealed that the late identified group had better scores on spelling than the early identified group, but still lower scores than the control group. The current study hypothesized that all students with dyslexia fall behind in spelling, but the group with a late diagnosis showed less difficulties with spelling. In this study, the results for the difference between the two groups with dyslexia are in line with the study of Leach et al. (2003), between the control group and late diagnosed students however, this study found no significant difference. The following paragraph will elaborate on the possible causes of the performance differential between spelling and reading, which in turn could give a better understanding of the difference in performing between the two groups with dyslexia.
That reading and spelling differs from each other is not surprising, because for reading, both speed and accuracy are important, whereas for spelling only accuracy is necessary. The group with an early diagnosis of dyslexia had difficulties with reading and spelling, as such, both speed and accuracy in reading and spelling were difficult for this group. The group with a late diagnosis of dyslexia only had difficulties with reading, so these students may have more trouble with speed than with accuracy. Moreover, for reading recognition of words or letter patterns is necessary, whereas for spelling, the active production of letters and words to write a word correctly is needed (Vaessen, 2010). To spell words correctly requires more specific production of letters and words than is required for
reading. Spelling requires knowledge of phoneme-grapheme connections, spelling rules, morphological rules, or word-specific knowledge (Vaessen, 2010). Reading is based on quick and accurate recognition of the connection between, and understanding of written words (Ouellette, 2006). These findings indicated that spelling is more difficult than reading, but the current study only found that the early diagnosed students had difficulties with spelling, whereas the late diagnosed students did not find it more difficult. Considering the difference, it can be cautiously stated that students with a late diagnosis only attract attention after elementary school because they only lag behind in reading, not in spelling.
In the domain of reading comprehension, no significant differences were found among the three groups. This finding is different than expected based on the simple view of reading, which assumes that language comprehension and decoding are necessary attributes in understanding the meaning of the text (Tunmer & Champman, 2012). As students with dyslexia have more difficulties with decoding, it could be possible that they focus more on decoding, and therefore have less consideration for comprehension of what they are reading. Muter, Hulme, Snowling, and Stevenson (2004), and Storch and Whitehurst (2002) found in their studies that for word reading ability, a person’s phonological awareness is a predictor, whereas for reading comprehension vocabulary knowledge is more important. Vocabulary skills are essential for understanding a text (Naiton, 2009). Students with and without dyslexia had no difficulties understanding a text, which might indicate strong vocabulary skills, helping them to comprehend texts. To understand a text, it is essential that students with and without dyslexia can read – adequate word reading however, won’t ensure that good reading comprehension will develop (Cain, 2010). There are aspects aside from word reading and vocabulary that influence reading comprehension, such as morphology, sentence comprehension, and discourse skills (Cain, 2010). All of these aspects influenced the performance of reading comprehension of the students with dyslexia in a positive way; in the current study, the students with dyslexia demonstrated well-developed reading comprehension. Reading comprehension is not included within the diagnosis of dyslexia, and although students with dyslexia experience difficulties with reading, they have no difficulties in understanding the meaning of a text. While this is an interesting finding the performance in reading comprehension of high school students is not well researched, making this interesting area for further research.
Reading comprehension has been studied by Leach et al. (2003), the difference however, is that within the current study, all students had an official diagnosis of dyslexia, whereas Leach et al. (2003) the sample selection was based on difficulties in reading. The
researchers found that the late identified group was heterogeneous and could be divided into three groups: a group with word-level processing deficits in combination with adequate comprehension skills, a group with weak comprehension skills accompanied by adequate lower-level skills, and a group with both kinds of difficulties.The study of Leach et al. (2003) found difficulties within reading comprehension, whereas the current study found difficulties only in reading and spelling. The students with dyslexia had no difficulties with reading comprehension. Although the study of Leach et al. (2003) and the current study focused on different groups, the group with deficits in decoding in combination with adequate comprehension skills which were also found in the study of Leach et al. (2003) are in line with the findings of the current study.
The last finding of the study concerned self-confidence: although students with dyslexia may have more difficulties in reading and spelling, which in turn might influence their self-confidence as compared to students without dyslexia, this study found no significant differences in self-confidence among the groups. We also found very low and non-significant correlations between self-confidence, and reading and spelling outcomes. The scale used in measuring self-confidence was created for the purpose of this study, so it’s reliability was been checked and was found to be high. Although the scale was reliable, no differences were found among the groups. Besides the questionnaire, questions were also asked reflecting a more personal view of students’ self-confidence. These questions revealed an interesting pattern: while all of the students compared their results with classmates, and the dyslexic groups read more slowly than the control group, the group with an early diagnosis was more irritated with reading more slowly than the group with a late diagnosis. These outcomes had to be interpreted carefully because the reliability of these questions was unacceptable low. Gunnel Ingesson (2007) found in interviews with dyslexic students that the majority only felt the influence of dyslexia on specific activities, such as reading and writing, but no difficulties were experienced in every other activities. The questionnaire in the current study is based on the students’ reading and writing abilities, but all of the students scored broadly the same on feelings about these abilities. These results were unexpected and different from other research (Gunnel Ingesson, 2007). The possible reason of the differences in findings is that Gunnel Ingesson (2007) used personal interviews, whereas the current study used a questionnaire. Self-confidence is someone’s personal interpretation of their abilities; students are able to provide more personal expression about their self-confidence in an interview than in a questionnaire.
Furthermore, Elbro’s study (2010) made a distinction between dyslexia as a disability and dyslexia as a handicap. This study found that adults with a large vocabulary had a higher risk of becoming more aware of their reading handicap in school. Within this study, most students had education level HAVO (general secondary education) or VWO (preparatory scientific education); because these are higher levels than VMBO (pre-vocational education), the students’ vocabulary was expected to be greater, and they might become more aware of their dyslexia as a handicap. However, this study found no differences among the groups in terms of self-confidence. Self-confidence is not frequently highlighted in research, so more studies on this subject among high school students are necessary to provide additional insights on the effects of self-confidence levels on their reading and writing performance.
The current study differs from other studies in that it is based on a sample of Dutch high school participants, and makes a distinction between students with an early and a late diagnosis of dyslexia. Despite these additions to existing literature, this study has several limitations. The first is that the students were mostly HAVO/VWO students; these students may have had higher vocabulary skills or are more intelligent as compared to VMBO students, influencing their scores on reading, spelling and reading comprehension (Elbro, 2010). This could also influence the spread of the scores of the total group. As well as for self-confidence, the scores might be different. As Elbro (2010) illustrated, there is a possibility that the social context and education of the participating students can influence their perception of themselves as dyslexic. It is possible that VMBO students could interpret their education as inferior and therefore view their dyslexia from a different perspective than HAVO/VWO students. The different interpretation on the part of VMBO students about their abilities could have influenced the scores on self-confidence. If the number of VMBO students was higher than it might have been possible to find significant scores for self-confidence. Moreover, including more VMBO students would better represent Dutch society, because the number of VMBO students is larger than the number of HAVO/VWO students.
The second limitation is that all of the participants were teenagers and volunteers, and some did not attend the meetings or forgot to fill in the questionnaire. If all participants had filled everything in correctly and with greater effort, the number of missing variables would be lower, especially concerning the questionnaire. Which would have influenced the scores for self-confidence. This study had some limitations that need more attention in further research. To provide a better view of high school students with dyslexia a large-scale, long-term investigation that could help conforming the findings of this study, and further expand on the effects of dyslexia on high school performance.
To conclude, the group with an early diagnosis and the group with a late diagnosis of dyslexia only differ in their performance in spelling, and not on the other variables. This shows that although the group with a late diagnosis has difficulties in reading, they do not make more mistakes or dropout in spelling, whereas the group with an early diagnosis has difficulties with and dropout on both reading and spelling. Another surprising result was found with regard to reading comprehension: whereas the two groups with dyslexia had difficulties with decoding, they had no trouble understanding the meaning of a text. This study examined whether the difference between the group with an early diagnosis and a group with a late diagnosis affected reading, spelling, reading comprehension, and self-confidence, in order to better understand how to help them. Based on the results of this study, reading remains important for both students with an early and a late diagnosis of dyslexia, as for the early group however, spelling needs to be taken into account as well. Despite these difficulties, students with dyslexia had no difficulties with understanding the meaning of a text, nor with their self-confidence.
As a final point, the results of this study imply that in terms of reading, additional attention of teachers is required, because all students with dyslexia in this study struggle with reading. Spelling is difficult for the group with an early diagnosis of dyslexia but not for the group with a late diagnosis of dyslexia, so additional attention in this sense should be aimed towards the early group. Although students with dyslexia have reading difficulties, they have no difficulties with understanding a text, so reading comprehension should not need to be an important aspect of additional help that students with dyslexia receive. Regarding self-confidence, no differences were found among the groups, but it should still be kept in mind that difficulties with self-confidence could be different if the sample within this study was larger. Further research is required on the appropriate form of attention in class. Regarding the implications for educational practice, the findings suggest that extra help in school for students with dyslexia should focus on improving their reading abilities, and should maintain awareness of their spelling, reading comprehension, and self-confidence.
References
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental
disorders: DSM-5 (5th ed.). Washington, DC: American Psychiatric Publishing Inc.
Blomert L. (2005). Dyslexie in Nederland: Theorie, praktijk en beleid. Amsterdam: Nieuwezijds.
Van den Bos, K. P. H. C., Lutje Spelberg, A. J. M., Scheepstra, & de Vries, J. R. (1994). De
klepel, een test voor de leesvaardigheid van pseudowoorden. Verantwoording, handleiding, diagnostiek en behandeling. Nijmegen: Berkhout.
Brus, B. T., & Voeten, M. J. M. (1999). Een minuut test (translation: One minute test). Lisse: Swets & Zeitlinger.
Burden, R. (2008). Is dyslexia necessarily associated with negative feelings of self-worth? A review and implications for future research. Dyslexia, 14(3), 188-196.
doi:10.1002/dys.371
Cain, K. (2010). Reading development and difficulties. BPS Blackwell, John Wiley & Sons. Carroll, J. M., & Iles, J. E. (2006). An assessment of anxiety levels in dyslexic students in
higher education. British Journal of Educational Psychology, 76(3), 651-662. doi:10.1348/000709905X66233
Catts, H. W., Compton, D., Tomblin, J. B., & Bridges, M. S. (2012). Prevalence and nature of late-emerging poor readers. Journal of educational psychology, 104(1), 166-181. doi:10.1037/a0025323
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Elbro, C. (2010). Dyslexia as disability or handicap: When does vocabulary matter? Journal
of learning disabilities, 43(5), 469-478. doi:10.1177/0022219409357349
Evers, A., Egberink, I. J. L., Braak, M. S. L., Frima, R. M., Vermeulen, C. S. M., & Van Vliet-Mulder, J. C. (2009–2012). COTAN documentatie. Amsterdam, Netherlands: Boom.
Examenblad. (2016) Retrieved on 21 May 2016 from:
http://www.examenblad.nl/examen/nederlands-vwo-2/2016
Gunnel Ingesson, S. (2007). Growing up with dyslexia: Interviews with teenagers and young adults. School psychology international, 28(5), 574-591. doi:10.1177
/0143034307085659
Harel, T., & Schaap, T. (1981). De samenhang van spellingsfouten. Amsterdam: Universiteit van Amsterdam.
Hoy, C., Gregg, N., Wisenbaker, J., Manglitz, E., King, M., & Moreland, C. (1997). Depression and anxiety in two groups of adults with learning disabilities. Learning
Disability Quarterly, 20(4), 280-291. doi:10.2307/1511226
Hulme, C., & Snowling, M. J. (2016). Reading disorders and dyslexia. Current Opinion in
Pediatrics, 28(6), 731-735. doi:10.1097/MOP.0000000000000411
KPC groep (2005). Signaleringsinstrument Protocol Dyslexie Voortgezet Onderwijs:
Handeleiding [herziene versie handleiding: Wonderlijke weer] Retrieved on 25 March from: http://masterplandyslexie.nl/public/files/documenten/Handleiding_
Signaleringsinstrument.pdf
Kuijpers, C., Van der Leij, A., Been, P., Van Leeuwen, T., Keurs, M. T., Schreuder, R., & Van den Bos, K. P. (2003). Leesproblemen in het voortgezet onderwijs en de volwassenheid. Pedagogische Studiën, 80(4), 272-287. Retrieved on 1 June 2016 from http://hdl.handle.net/11370/26baaeee-066c-4f24-a12a-47c88586ab99 Lawrence, D. (2006). Enhancing self-esteem in the classroom. London, Paul Chapman
Publishing.
Leach, J. M., Rescorla, L., & Scarborough, H. S. (2003). Late-emerging reading disabilities.
Journal of Educational Psychology, 95(2), 211-224. doi:10.1037/0022-0663.95.2.211
McNulty, M. A. (2003). Dyslexia and the life course. Journal of learning disabilities, 36(4), 363-381. doi:10.1177/00222194030360040701
Muter, V., Hulme, C., Snowling, M., & Stevenson, J. (2004). Phonemes, rimes, vocabulary and grammatical skills as foundations of early reading development: Evidence from a longitudinal study. Developmental Psychology, 40(5), 81-665. doi:10.1037/0012-1649.40.5.665
Nation, K. (2009). Reading comprehension and vocabulary: What’s the connection? In R. K. Wagner, C. Schatschneider, C. Phythian-Sence (Eds.), Reading comprehension and vocabulary. Beyond decoding (pp. 176-194). New York, The Guildford Press. Ouellette, G. P. (2006). What's meaning got to do with it: The role of vocabulary in word
reading and reading comprehension. Journal of educational psychology, 98(3), 554. doi: 10.1037/0022-0663.98.3.554
Perry, P. (2011). Concept analysis: Confidence/self-confidence. Nursing forum,
46(4), 218-230. doi:10.1111/j.1744-6198.2011.00230.x
Retelsdorf, J., & Köller, O. (2014). Reciprocal effects between reading comprehension and spelling. Learning and Individual Differences, 30, 77-83. doi: http://dx.doi.org/ 10.1016/j.lindif.2013.11.007
SDN, De Jong, P. F., De Bree, E. H., Henneman, K., Kleijnen, R., Loykens, E. H. M., Rolak, M., ….. & Wijnen, F. N. K. (2016). Dyslexie: diagnostiek en behandeling. Brochure
van de Stichting Dyslexie Nederland. Retrieved on 28 March 2017 from
http://www.stichtingdyslexienederland.nl/media/1643/ dyslexie_diagnostiek%20en%20behandeling_2016.pdf
Sontag, L., & Donker, M. (2012). Dyslexie en dyscalculie in het voortgezet onderwijs.
Herhaalde meting. (Research Report). ´s-Hertogenbosch, Netherlands: KPC Groep.
Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading: Evidence from a longitudinal structural model, Developmental Psychology,
38(6), 47-934. doi:10.1037/0012-1649.38.6.934
Torppa, M., Eklund, K., van Bergen, E., & Lyytinen, H. (2015). Late-emerging and resolving dyslexia: A follow-up study from age 3 to 14. Journal of abnormal child psychology,
43(7), 1389-1401. doi:10.1007/s10802-015-0003-1
Tunmer, W. E., & Chapman, J. W. (2012). The simple view of reading redux vocabulary knowledge and the independent components hypothesis. Journal of learning
disabilities, 45(5), 453-466. doi:10.1177/0022219411432685
Vaessen, A. A. (2010). Cognitive dynamics of fluent reading and spelling development. (Doctoral dissertation). Retrieved on 16 February 2017 from:
http://digitalarchive.maastrichtuniversity.nl/fedora/get/guid:d2f495c6-5db5-4702-8124-416fe2ca9102/ASSET1. (University of Maastricht)
