Evaluating a phonics-based reading intervention for children with developmental language disorder – A single-blinded randomized controlled trial

ERCT Criteria Breakdown

• Level 1 Criteria

  • C

    Class-level RCT

    • Participants were randomized at the individual child level rather than by class (and the intervention is not described as one-to-one tutoring), so the class-level RCT requirement is not satisfied.
    • "When entering the study, the children with DLD were randomized, based on school-grade and sex, into three subgroups (strata), the GL-intervention group or one out of two control groups: one active control group playing a math computer game (Math group), and one passive control group undertaking teaching as usual (TAU group, see Figure 1)." (p. 5)
    • Relevant Quotes: 1) "Sixty-seven children enrolled in Grades 0–4 were randomized into three groups: one intervention group playing GL, one active control group playing a math game, and one control group that received teaching as usual." (Abstract) 2) "The participants with DLD were recruited from nine school language units in Sweden, where, in total, 209 children with DLD were enrolled." (p. 4) 3) "These schools have small classes for a maximum of 8– 10 children with DLD." (p. 4) 4) "When entering the study, the children with DLD were randomized, based on school-grade and sex, into three subgroups (strata), the GL-intervention group or one out of two control groups: one active control group playing a math computer game (Math group), and one passive control group undertaking teaching as usual (TAU group, see Figure 1)." (p. 5) 5) "Participants were then randomized within each stratum to ensure equal representation (block size randomization) in all intervention groups." (p. 5) Detailed Analysis: ERCT Criterion C requires randomization at the class level (or stronger, such as school-level) to reduce cross-condition contamination within classrooms and teacher practices. In this paper, the allocation is explicitly described as randomizing "the children" and then randomizing "within each stratum," which indicates an individual-level RCT rather than class-level assignment. The study context (nine language units with small classes) increases the plausibility that children and staff could interact across study arms within the same unit, which is precisely what Criterion C is meant to reduce. The exception for one-to-one tutoring/personal teaching does not clearly apply here because the intervention is a computer game played in school with a teacher "available in the room," not described as individualized tutoring as the intervention design itself. Criterion C is not met because randomization is at the individual child level, not at the class (or school) level, and no tutoring exception is clearly applicable.
  • E

    Exam-based Assessment

    • The study includes a widely recognized standardized test of decoding (TOWRE), satisfying the exam-based assessment requirement.
    • "Word decoding and nonsense word decoding were measured with the Swedish version of the Test of Word Reading Efficiency, TOWRE (Lyxell et al., 2009)." (p. 7)
    • Relevant Quotes: 1) "Word decoding and nonsense word decoding were measured with the Swedish version of the Test of Word Reading Efficiency, TOWRE (Lyxell et al., 2009)." (p. 7) 2) "TOWRE consists of two subtests that assess word decoding and nonsense word decoding." (p. 7) 3) "Raw scores, but also standardized scores based on new Swedish reference data (see below for details), were reported." (p. 7) Detailed Analysis: ERCT Criterion E requires using standardized, widely recognized exam-based (standardized) assessments rather than relying only on bespoke or intervention-aligned measures. While the paper also uses a GraphoLearn-based decoding task as its primary outcome, it clearly includes TOWRE, a widely used standardized decoding assessment, as a secondary outcome. This constitutes a standardized assessment instrument that is independent of the specific game content and is not newly created for this study. Therefore, the study satisfies Criterion E by including a recognized standardized assessment (TOWRE), even though not all outcomes are standardized exams. Criterion E is met because the study uses TOWRE, a widely recognized standardized reading test, as an outcome measure.
  • T

    Term Duration

    • The study measures outcomes through follow-up over about 16 weeks from baseline, which is approximately one full academic term.
    • "The study adopted a longitudinal design ranging over 16 weeks (see Figure 1)." (p. 5)
    • Relevant Quotes: 1) "The duration of the intervention was 5 weeks." (Abstract) 2) "Decoding ability was measured at baseline, post-intervention, and at follow-up 10–12 weeks after the intervention." (Abstract) 3) "The study adopted a longitudinal design ranging over 16 weeks (see Figure 1)." (p. 5) 4) "After the first baseline assessment (T1), an intervention period of 5 weeks was initiated." (p. 5) Detailed Analysis: ERCT Criterion T requires outcomes to be measured at least one full academic term after the intervention begins (typically about 3–4 months). The paper describes a 5-week intervention and a follow-up occurring 10–12 weeks after the intervention, and it summarizes the total longitudinal design as spanning 16 weeks from baseline (T1) to follow-up (T3). Sixteen weeks is approximately 3.5–4 months, which aligns with a typical term-length tracking window. Criterion T is met because outcome measurement extends to an approximately 16-week follow-up window from baseline, meeting the minimum term-duration requirement.
  • D

    Documented Control Group

    • The paper clearly defines both control conditions and reports baseline comparability across groups, satisfying the documented control group requirement.
    • "Sixty-seven children enrolled in Grades 0–4 were randomized into three groups: one intervention group playing GL, one active control group playing a math game, and one control group that received teaching as usual." (Abstract)
    • Relevant Quotes: 1) "Sixty-seven children enrolled in Grades 0–4 were randomized into three groups: one intervention group playing GL, one active control group playing a math game, and one control group that received teaching as usual." (Abstract) 2) "When entering the study, the children with DLD were randomized... into three subgroups (strata), the GL-intervention group or one out of two control groups: one active control group playing a math computer game (Math group), and one passive control group undertaking teaching as usual (TAU group, see Figure 1)." (p. 5) 3) "The TAU group undertook teaching as usual, that is, classroom-based teaching in small groups that followed the structure of the Swedish national curriculum." (p. 6) 4) "There were no differences between the three groups at T1 on age, general language ability, non-verbal IQ, non-word repetition, or any of the reading measures." (p. 8) Detailed Analysis: ERCT Criterion D requires sufficiently detailed documentation of the control group(s), including what they received and evidence that baseline characteristics are reported to support comparability checks. The paper describes two control groups (active math-game control and passive teaching-as-usual control) and provides a concrete description of TAU. It also reports that there were no baseline differences among groups at T1 across key demographic, language, and reading measures, with descriptive statistics provided in the paper. Criterion D is met because control conditions are clearly defined and baseline group characteristics/comparability are documented.

• Level 2 Criteria

  • S

    School-level RCT

    • Schools/language units were recruitment sites, but randomization was conducted at the child level rather than by school, so the school-level RCT criterion is not satisfied.
    • "When entering the study, the children with DLD were randomized, based on school-grade and sex, into three subgroups (strata)..." (p. 5)
    • Relevant Quotes: 1) "The participants with DLD were recruited from nine school language units in Sweden..." (p. 4) 2) "When entering the study, the children with DLD were randomized, based on school-grade and sex, into three subgroups (strata)..." (p. 5) 3) "Participants were then randomized within each stratum to ensure equal representation (block size randomization) in all intervention groups." (p. 5) Detailed Analysis: ERCT Criterion S requires that the unit of randomization is the school/site (i.e., schools are randomized to conditions), which better reflects real-world rollouts and avoids within-school spillovers. Although the sample is recruited from nine language units, the paper repeatedly describes randomizing the children (and then randomizing within strata). There is no statement that entire schools/language units were assigned to study conditions. Criterion S is not met because allocation is not at the school level.
  • I

    Independent Conduct

    • GraphoLearn is described as an externally developed program rather than one designed by the study authors, supporting independent conduct relative to the intervention’s designers.
    • "The GL game is a computer-based reading intervention, originally from Finland (Lyytinen et al., 2007, 2009)..." (p. 4)
    • Relevant Quotes: 1) "The GL game is a computer-based reading intervention, originally from Finland (Lyytinen et al., 2007, 2009), translated from Finnish Swedish into standard Swedish (Nakeva von Mentzer, 2014)." (p. 4) 2) "In each school, a special needs teacher or a speech-language pathologist was responsible for implementing the intervention." (p. 6) 3) "The intervention fidelity was closely monitored by the research team using in-game data from the GL server to ensure that the participants were playing GL according to the study design, and feedback was given to the school when needed." (p. 6) Detailed Analysis: ERCT Criterion I requires that the evaluation is conducted independently of the people who designed the intervention, to reduce bias from testing one's own product. The paper explicitly describes GraphoLearn as "originally from Finland" and cites external publications for its development and adaptation, suggesting the intervention itself is not authored/designed by this paper's author team. The delivery is carried out by school staff, not by the authors, even though the research team monitors fidelity. The paper does not present evidence that the authors are the GraphoLearn designers; rather, it frames the intervention as externally developed and implemented by practitioners in schools. Criterion I is met because the intervention is described as externally developed (not designed by the authors) and implemented by school personnel, supporting independent conduct relative to the intervention’s designers.
  • Y

    Year Duration

    • The study spans about 16 weeks from baseline to follow-up, which is far shorter than 75% of an academic year.
    • "The study adopted a longitudinal design ranging over 16 weeks (see Figure 1)." (p. 5)
    • Relevant Quotes: 1) "The study adopted a longitudinal design ranging over 16 weeks (see Figure 1)." (p. 5) 2) "After the first baseline assessment (T1), an intervention period of 5 weeks was initiated." (p. 5) 3) "Decoding ability was measured at baseline, post-intervention, and at follow-up 10–12 weeks after the intervention." (Abstract) Detailed Analysis: ERCT Criterion Y requires outcomes to be measured at least 75% of an academic year after the intervention begins. The paper explicitly states the design spans 16 weeks from baseline through follow-up. This is substantially less than 75% of a typical 9–10 month school year, and there is no claim of year-long outcome measurement for this cohort. Criterion Y is not met because the tracking period is only about 16 weeks rather than most of an academic year.
  • B

    Balanced Control Group

    • The active control matches the game dosage, but the TAU condition is not described as receiving comparable time and resources, and the paper does not clearly state that added resources/time are the treatment variable.
    • "The intervention group was instructed to play a minimum of 20 sessions of the GL game, for a minimum of 20 min and a maximum of 30 min per session..." (p. 5)
    • Relevant Quotes: 1) "The intervention group was instructed to play a minimum of 20 sessions of the GL game, for a minimum of 20 min and a maximum of 30 min per session, using headphones." (p. 5) 2) "The Math group followed the same procedure as the GL group, namely 20 sessions during the same period." (p. 5) 3) "The TAU group undertook teaching as usual, that is, classroom-based teaching in small groups that followed the structure of the Swedish national curriculum." (p. 6) 4) "This study aimed to investigate the effect of a phonics-based computer game, GraphoLearn (GL)..." (p. 3) Detailed Analysis: ERCT Criterion B requires that time and resources are balanced between intervention and control conditions, unless the study explicitly frames added time/resources as the treatment variable being tested. The paper clearly specifies a structured "dosage" for the GraphoLearn group (20 sessions, 20–30 minutes, headphones) and states that the active control (Math game) follows the same procedure, which supports balance for that comparison. However, the paper includes a TAU group described only as normal classroom-based teaching. It does not clearly state whether the game sessions replaced existing instructional time (resource-neutral) or were added on top of usual instruction (extra time/resource). It also does not explicitly frame "adding extra instructional time or extra resources" as the treatment variable; instead, it frames the intervention as the specific GraphoLearn program. Because the TAU condition is not documented as receiving comparable time-on-task, device access, supervision, or an equivalent substitute activity during the same minutes, the balance of educational inputs across all conditions cannot be confirmed. Criterion B is not met because equivalence of time and resources is not clearly documented for the TAU group, and added resources are not explicitly framed as the treatment variable.

• Level 3 Criteria

  • R

    Reproduced

    • No independent replication of this specific RCT could be identified, and the paper itself frames the trial as the first RCT of this kind in this population.
    • "To our knowledge, this is the first randomized controlled trial investigating the impact of a phonics-based intervention method in children with DLD." (Abstract)
    • Relevant Quotes: 1) "To our knowledge, this is the first randomized controlled trial investigating the impact of a phonics-based intervention method in children with DLD." (Abstract) Detailed Analysis: ERCT Criterion R requires evidence that an independent research team has replicated the study in a different context and published it in a peer-reviewed venue. The paper explicitly positions itself as the first RCT of this type in children with DLD, which suggests that a replication of this specific trial is not established within the paper. In addition, an external search (as of 2026-02-13) did not identify a peer-reviewed, independent replication that explicitly reproduces this exact trial design with GraphoLearn versus math-game and TAU controls in Swedish language units. Criterion R is not met because there is no evidence of an independent peer-reviewed replication of this specific RCT.
  • A

    All-subject Exams

    • The study assesses reading/decoding outcomes rather than standardized exam outcomes across all main school subjects, so the all-subject exams criterion is not met.
    • "Three measures of reading were used." (p. 7)
    • Relevant Quotes: 1) "Three measures of reading were used." (p. 7) 2) "The primary outcome was measured using the GL decoding task..." (p. 7) 3) "Word decoding and nonsense word decoding were measured with the Swedish version of the Test of Word Reading Efficiency, TOWRE (Lyxell et al., 2009)." (p. 7) Detailed Analysis: ERCT Criterion A requires standardized exam-based assessment across all main subjects taught at the relevant educational level (not only the intervention’s focal subject), to detect possible trade-offs. This study focuses on reading/decoding outcomes and describes only reading-related measures. It does not measure outcomes in other main subjects (for example, mathematics, science, or broader language-arts outcomes beyond decoding), and it is not presented as a specialized upper-secondary/vocational intervention where a narrow subject set would be justified. Criterion A is not met because outcome measurement is limited to reading/decoding and does not cover all main school subjects using standardized exams.
  • G

    Graduation Tracking

    • The study ends after a short follow-up (weeks) and does not track students until graduation; additionally, ERCT specifies that failing year-duration (Y) implies failing graduation tracking (G).
    • "Decoding ability was measured at baseline, post-intervention, and at follow-up 10–12 weeks after the intervention." (Abstract)
    • Relevant Quotes: 1) "Decoding ability was measured at baseline, post-intervention, and at follow-up 10–12 weeks after the intervention." (Abstract) 2) "The study adopted a longitudinal design ranging over 16 weeks (see Figure 1)." (p. 5) Detailed Analysis: ERCT Criterion G requires tracking participants until graduation from the relevant educational stage. ERCT also specifies that if Criterion Y (year duration) is not met, Criterion G cannot be met. The paper’s design spans 16 weeks and ends at a follow-up 10–12 weeks after the intervention. There is no mention of tracking participants through completion of Grade 4 or any later graduation milestone, and the duration is far short of a school-year, let alone until graduation. An external search (as of 2026-02-13) did not identify a follow-up publication by the same author team that tracks this cohort to graduation. Criterion G is not met because the study does not track to graduation and the prerequisite year-duration criterion is not met.
  • P

    Pre-Registered

    • The trial is registered (NCT05295472), but registry dates indicate it was first posted after the study start and after data collection had begun, so it is not pre-registered.
    • "First Posted (Actual) March 25, 2022" (ClinicalTrials.gov record for NCT05295472)
    • Relevant Quotes: 1) "The trial was also registered at https://clinicaltrials.gov/ with registration number NCT05295472." (p. 3) 2) "Data was collected during 18 months, from September 2021 to February 2023." (p. 5) 3) "Study Start (Actual) August 1, 2021" (ClinicalTrials.gov record for NCT05295472) 4) "First Posted (Actual) March 25, 2022" (ClinicalTrials.gov record for NCT05295472) Detailed Analysis: ERCT Criterion P requires that the study protocol is pre-registered before the study begins (i.e., registration date must precede the start of data collection). The paper reports that the trial was registered on ClinicalTrials.gov under NCT05295472, and it reports that data collection ran from September 2021 to February 2023. Checking the trial registry record for NCT05295472 shows a study start date of August 1, 2021 and a first posted date of March 25, 2022. This indicates that registration was posted after the study started and after data collection had already begun (given the paper’s September 2021 start of data collection). Criterion P is not met because the registry posting date occurs after the study start and after data collection began, so the protocol was not pre-registered.