Reading delay and difficulty: What is the role of risk for intervention effectiveness?

Reading delay and difficulty: What is the role of risk for intervention effectiveness?

First Author: Jan Frijters -- Brock University
Keywords: Reading disability, Intervention studies, Early intervention, socioeconomic status, At Risk Students
Abstract / Summary: 

Risk for reading delay and/or difficulty ranges from early deficits in preliteracy and language skills necessary for reading development to broader social and economic factors. The relationship between individual risk factors and reading skill has had good coverage in extant research. However, there has been much less focus on how risk, broadly defined, intersects with evidence-based interventions aimed at addressing reading failure and/or building reading skill. The four papers in the proposed symposium address risk along two dimensions. The first dimension is early risk, with Rigmor-Walgermo, Piasta, and Mirza reporting on interventions preselecting participants based on early screeners. The second dimension is broader risk (i.e., social, economic,  or educational), with Greenberg and Mirza’s projects contain interventions conducted in populations at broader risk. The discussant will draw on 20 years of experience in reading intervention to address the ideas of cumulative risk, risk burden, and compensatory vs. restitutive effects of intervention in relation to interpreting intervention outcomes.

Symposium Papers: 

Effects of a small-group, emergent literacy intervention for preschoolers at risk for reading difficulties when implemented under routine conditions by classroom teachers and community aides

First Author/Chair:Shayne Piasta -- The Ohio State University
Additional authors/chairs: 
Jessica A.R. Logan; Cynthia M. Zettler-Greeley; Laura L. Bailet; Kandia Lewis; Leiah J. Groom

Purpose: We examined the efficacy of the Nemours BrightStart! intervention for preschoolers at risk for later reading difficulties when used under routine conditions. Prior research, conducted by the developers under highly controlled conditions, demonstrated effects on children’s phonological awareness and print knowledge skills (Bailet et al., 2009, 2013; Zettler-Greeley et al., 2018). We extend this work to examine effects when implemented by classroom teachers or adults affiliated with a local kindergarten-readiness initiative (i.e., community aides).

Method:  We recruited 281 children from 98 preschool classrooms, all of whom were at risk based on the Get Ready to Read screener.  We randomly assigned classrooms to teacher-implemented, community aide-implemented, or business-as-usual control conditions.  Children assigned to intervention conditions experienced weekly lessons implemented by their teachers or visiting community aides.  All children completed pretest and posttest assessments. We analyzed data using multilevel residualized change models.

Results:  Intent-to-treat analyses indicated no significant differences between children assigned to the intervention conditions versus control on phonological awareness, print knowledge, language, or emergent writing skills.  However, fidelity data showed that not all children received the intervention as assigned; 50% in the teacher-implemented condition received ≤ 3 lessons, whereas 30% in the community aide-implemented condition received ≤ 10 lessons. To account for dosage differences, we will use an instrumental variables approach; these results will also be presented.

Conclusions:  Findings mirror previous work showing reduced effects when using interventions under real-world routine conditions, highlight the importance of evaluating interventions under such conditions, and have implications for scaling up intervention effor

First Grade Reading Intervention and the dynamics of interest for reading and reader self-concept

First Author/Chair:Bente Rigmor-Walgermo -- University of Stavanger
Additional authors/chairs: 
Jan C. Frijters; Oddny Judith Solheim

Purpose: The development of reading skill and motivation for reading is connected even at the earliest stages of reading development. Consequently, already within the first year of formal reading instruction students with poor early reading skills tend to have a weaker reader self-concept and to make more use of task-avoidance strategies than their peers. The main aim of the present study was to investigate whether participation in an intensive first-grade reading intervention was associated with changes in student´s reader self-concept and interest for reading.

Method: The sample consisted of 1,141 students (mean age 6.15), of whom 212 were identified as being at risk of reading difficulties (RD) at school entry. All students were measured at the beginning of Grade 1 and at the end, with the at-risk students randomly assigned (full RCT) to receive a control condition or an evidence-based intervention.

Results: The results show that emergent readers considered to be at risk of RD had a significantly weaker reader self-concept than their not-at-risk peers at school entry with a substantial effect size (d = .38). No associations between risk-status and level of interest were found. Further, no main effect of intervention was observed on either interest nor reader self-concept measures at the end of Grade 1. However, an effect of the intervention was found on the interplay between interest and self-concept, which varied depending on risk status and intervention assignment.

Conclusions: While interest was a crucial driving force for self-concept for students in the control-condition, this was not the case for those who participated in the intervention- and for students originally not at risk. These findings are discussed up against the design and content of early reading intervention programs.

EmpowerTM Reading Intervention: An experimental approach to remediate reading disabilities among at-risk population

First Author/Chair:Amna Mirza -- Brock University
Additional authors/chairs: 
Maureen Lovett; Karen Steinbach; Joan Bosson-Heenan; Jeffrey Gruen; Jan C. Frijters

Purpose: The cardinal features of reading disabilities are significant impairments in the development of reading accuracy, fluency, comprehension and spelling skills. Rates of RD are higher within high-poverty urban schools, low familial SES and empirically underrepresented groups, African Americans and Hispanics. Overall, children with RD show weak awareness of spoken words’ sound-structure and ability to make phonological representations in memory, but there is little evidence that these deficits are inherent to any of the risk groups listed above. Therefore, targeting these skills should compensate for the reading-specific deficits, regardless of risk factors.

Method: This study utilized EmpowerTM Reading as a remedial effort within a population that shared all risk-factors mentioned above. 91, grade 2 children from New Haven Lexinome Project-NHLP received 60 minutes of intervention for 110 hours. Children were also tested on five experimental measures at four time-points to test the effectiveness of the implemented intervention.

Results: A growth curve analysis was conducted on letter-sound, sound-combination, keywords, transfer and challenge words. Children showed significant growth on sound-symbols and keywords. Growth line on the challenge and transfer words suggested that children’s developed skills to decode words they hadn’t learnt previously. When compared to a control-group drawn from non-intervention schools, substantial differences emerged on standardized word-reading measures. Post-hoc analyses indicated no difference in effectiveness for those with/without any of the risk factors.

Conclusions: Findings suggest that early identification and intervention with explicit instructions on core areas of reading can help with remediation of reading disabilities among children who are at-high risk of reading failures.

Cumulative and domain-specific risk influences on both skill and motivation change among adults with low literacy skills

First Author/Chair:Greenberg Daphne -- Georgia State University
Additional authors/chairs: 
Maureen Lovett; Arthur Graesser; Jan C. Frijters

Purpose: Adults with low reading skills have complex learning histories that are influenced by multiple risk factors. The cumulative burden on reading development includes educational, health, and economic risks. A multiple component intervention designed to address the multifaceted reading problems of adult learners was implemented, and potential efficacy on reading and motivation for reading was tracked across multiple risk types.

Method: 320 adults, reading between the 3rd and 7th grades, participated in research classes (N = 233) or their usual literacy classes (control, N = 87).  The intervention included teacher-led decoding, vocabulary, and comprehension instruction, followed by comprehension work using an intelligent tutoring system. Cumulative risk indices were formed via self-report instruments that encompassed 57 questions related to the three risk dimensions.

Results:  Significant intervention versus control effects were observed for decoding and motivation outcomes. Additionally, skill and motivation growth over the study duration was influenced by educational risk level (outcome: sense of competence), health risk (outcome: perceived difficulty), and economic risk (outcomes: nonword decoding; reading interest). In every case, those with the greatest levels of risk benefited most from intervention.

Conclusions: While cumulative risk levels across three dimensions had inconsistent relationships with reading skill and motivation at pretest, greater positive change was observed for those with the highest risk at time of intervention. Discussion will focus on the active elements of the intervention and compensation of specific risk factors through intervention, as evidenced by the specific pattern of risk-outcome effects.

Discussion of the intervention papers in terms of broadly determined risk concepts

First Author/Chair:DISCUSSANT Jan C. Frijters