Peer Assisted Learning Strategies (PALS-UK) Increases Reading Attainment, Oral Fluency and Comprehension: A Cluster Randomized Controlled Trial

ERCT Criteria Breakdown

• Level 1 Criteria

  • C

    Class-level RCT

    • Schools were randomized (cluster RCT), which meets and exceeds the class-level randomization requirement.
    • "All Year 5 pupils (9–10 years old, N = 4840, 49% female and 51% male) in 114 schools in England, took part in a two-armed, randomized controlled efficacy trial randomizing schools."
    • Relevant Quotes: 1) "All Year 5 pupils (9–10 years old, N = 4840, 49% female and 51% male) in 114 schools in England, took part in a two-armed, randomized controlled efficacy trial randomizing schools." (p. 1) 2) "We conducted a two-armed, pragmatic, cluster-randomized controlled efficacy trial. Schools" (p. 5) 3) "At randomization, 114 schools were assigned 1:1 to treatment and control groups." (p. 5) Detailed Analysis: Criterion C requires that randomization occur at the class level (or stronger, such as school-level) to reduce within-school/class contamination. The abstract explicitly states the trial is "randomizing schools" and the Methods section states "Schools" were assigned, with 114 schools assigned 1:1 to treatment and control. School-level randomization is stronger than class-level randomization, so it satisfies Criterion C. Final Summary: Criterion C is met because the unit of randomization was the school, which is stronger than class-level randomization.
  • E

    Exam-based Assessment

    • The outcomes use standardized, widely recognized assessments (New PiRA and WIAT-III UK-T) administered under standardized procedures.
    • "New PiRA is a series of termly UK curriculum-aligned standardized norm-referenced assessments."
    • Relevant Quotes: 1) "New PiRA is a series of termly UK curriculum-aligned standardized norm-referenced assessments." (p. 8) 2) "New PiRA was administered to the whole class by the classroom teacher in accordance with standardized instructions." (p. 8) 3) "Reading comprehension and oral reading fluency subtests from the WIAT-III UK-T (Wechsler, 2018) were administered individually by trained testers, in line with standardized instructions, including use of chronological age-related start/stop points and performance-related reverse rules." (p. 9) Detailed Analysis: Criterion E requires standardized exam-based assessment(s), not researcher-created tests tailored to the intervention. The paper explicitly identifies New PiRA as "standardized norm-referenced" and describes administration "in accordance with standardized instructions." It also uses WIAT-III UK-T subtests administered by trained testers "in line with standardized instructions." These descriptions indicate established standardized assessments and standardized administration procedures. Final Summary: Criterion E is met because the primary and key secondary outcomes are measured with standardized assessments (New PiRA; WIAT-III UK-T).
  • T

    Term Duration

    • Outcomes were measured from Summer 2022 (baseline) to Summer 2023 (endline), which is well beyond one academic term after intervention began.
    • "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022) and again at endline after pupils in the treatment group had finished the program (end of Year 5, Summer 2023)."
    • Relevant Quotes: 1) "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022) and again at endline after pupils in the treatment group had finished the program (end of Year 5, Summer 2023)." (p. 8) 2) "Schools were asked to deliver PALS-UK three times a week for 20 school weeks, from October 2022 to May/June 2023." (p. 7) Detailed Analysis: Criterion T requires that outcomes be measured at least one academic term (roughly 3–4 months) after the intervention begins. The paper indicates outcomes were measured at baseline in Summer 2022 and at endline in Summer 2023, after the program finished, and also documents a multi-month delivery period from October 2022 to May/June 2023. These timelines clearly exceed one academic term from intervention start to outcome measurement. Final Summary: Criterion T is met because outcomes were assessed at endline in Summer 2023 after an intervention period spanning multiple months.
  • D

    Documented Control Group

    • The control condition is explicitly described as business-as-usual, with clear sample sizes and baseline balance information reported.
    • "Control schools did not receive training and did not deliver PALS-UK to their pupils; they continued using their usual approaches to reading."
    • Relevant Quotes: 1) "The final analyzed sample included 53 treatment schools (N = 1907, 51% female and 49% male) and 50 business-as-usual control schools (N = 1721, 49% female and 51% male)." (p. 1) 2) "Control schools did not receive training and did not deliver PALS-UK to their pupils; they continued using their usual approaches to reading." (p. 5) 3) "Table 1. Distribution of baseline school and pupil characteristics by trial arm at randomization and analysis." (p. 6) Detailed Analysis: Criterion D requires that the control group be well documented, including what they received and baseline characteristics enabling comparison. The paper defines the control as "business-as-usual," states control schools did not receive training and did not deliver PALS-UK, and reports control sample sizes in the abstract. It also provides a baseline characteristics table by arm. Final Summary: Criterion D is met because the control group is clearly defined and described, with sample sizes and baseline characteristics reported.

• Level 2 Criteria

  • S

    School-level RCT

    • Schools (not classes or students) were randomized, satisfying the school-level RCT criterion.
    • "At randomization, 114 schools were assigned 1:1 to treatment and control groups."
    • Relevant Quotes: 1) "All Year 5 pupils (9–10 years old, N = 4840, 49% female and 51% male) in 114 schools in England, took part in a two-armed, randomized controlled efficacy trial randomizing schools." (p. 1) 2) "At randomization, 114 schools were assigned 1:1 to treatment and control groups." (p. 5) Detailed Analysis: Criterion S requires randomization at the school level. The abstract and Methods section explicitly state that schools were randomized, and specify that 114 schools were assigned 1:1 to treatment and control. Final Summary: Criterion S is met because the unit of randomization was the school.
  • I

    Independent Conduct

    • The paper indicates intervention development by study authors, but does not clearly document that evaluation and analysis were led independently from the intervention designers.
    • "This package of teacher training differs somewhat from the original PALS Grade 2–6 US version, as it has been specifically developed for the UK context (by the 1st and 5th Authors)."
    • Relevant Quotes: 1) "This package of teacher training differs somewhat from the original PALS Grade 2–6 US version, as it has been specifically developed for the UK context (by the 1st and 5th Authors)." (p. 7) 2) "We would also like to thank AlphaPlus and FFT Education (for supporting data collection and enumeration of the sample)..." (p. 20) Detailed Analysis: Criterion I requires clear documentation that the evaluation was conducted independently from the intervention designers. The paper explicitly states that key parts of the intervention package (teacher training) were developed by authors ("the 1st and 5th Authors"), indicating that intervention design and authorship overlap. While the Acknowledgments mention external organizations supporting data collection, the paper does not provide a clear statement that an independent evaluation team (separate from the intervention designers) led the trial conduct and analysis, nor does it clearly separate designer roles from evaluator roles within the study. Final Summary: Criterion I is not met because the paper does not clearly document independent conduct separate from the intervention designers.
  • Y

    Year Duration

    • The intervention and outcome tracking span from October 2022 through endline in Summer 2023, which is consistent with at least 75% of an academic year after intervention start.
    • "Schools were asked to deliver PALS-UK three times a week for 20 school weeks, from October 2022 to May/June 2023."
    • Relevant Quotes: 1) "Schools were asked to deliver PALS-UK three times a week for 20 school weeks, from October 2022 to May/June 2023." (p. 7) 2) "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022) and again at endline after pupils in the treatment group had finished the program (end of Year 5, Summer 2023)." (p. 8) Detailed Analysis: Criterion Y requires outcomes to be measured at least 75% of one academic year after the intervention begins. The paper describes delivery running "from October 2022 to May/June 2023" and endline outcome measurement at "end of Year 5, Summer 2023." In a typical England academic year, October to late May/June (and then endline in Summer) represents most of the school year following the intervention start. Although the paper also describes "20 school weeks" of delivery, the explicit calendar span and the endline timing indicate tracking from intervention start to endline over most of the academic year. Final Summary: Criterion Y is met because intervention start (October 2022) to endline (Summer 2023) spans roughly an academic year-length period.
  • B

    Balanced Control Group

    • The intervention resources (structured peer-reading sessions plus training/materials) are integral to the treatment being tested and are not framed as a separable, confounding add-on.
    • "Lessons took around 35 minutes (including setup) with activities completed in a set order..."
    • Relevant Quotes: 1) "Control schools did not receive training and did not deliver PALS-UK to their pupils; they continued using their usual approaches to reading." (p. 5) 2) "Lessons took around 35 minutes (including setup) with activities completed in a set order: Partner Reading (10 minutes), Retell (2 minutes), Paragraph Shrinking (10 minutes), and Prediction Relay (10 minutes)." (p. 7) 3) "Before classroom delivery, teachers attended a one day in-person training event (repeated in seven regional locations)." (p. 7) 4) "Hence, here we provided a selection of 20 age-appropriate contemporary fiction books per class, emphasizing that these should supplement existing book supplies and were not central to the programme." (p. 7) Detailed Analysis: Criterion B compares the nature, quantity, and quality of resources (time, materials, training/support) provided to intervention and control conditions, and asks whether any extra resources are either (a) balanced in the control condition, or (b) explicitly integral to the treatment being tested (i.e., the treatment is the full package, not merely a pedagogical idea). The intervention includes structured 35-minute sessions and teacher training. The control group is explicitly "business-as-usual" and did not receive PALS-UK training or deliver PALS-UK. Importantly, the paper does not describe PALS-UK as being delivered in addition to schooling time; it is described as a whole-class approach to be delivered by teachers, which plausibly substitutes for other reading instruction time within the timetable rather than adding extra instructional time overall. The teacher training and materials (including the supplemental class book set) are part of the implementation package being tested. Although the books are described as "not central to the programme," they are explicitly framed as supplementary to existing supplies, which reduces the likelihood that this is a large, separable resource confound. Final Summary: Criterion B is met because the key added inputs (structured sessions and training/materials) are integral to the intervention package being tested rather than an accidental resource imbalance.

• Level 3 Criteria

  • R

    Reproduced

    • No independent peer-reviewed replication by a different research team of this specific 2022/23 PALS-UK cluster RCT was found.
    • "The present study is a partial replication of that trial, and it represents the most rigorous test of the impact of PALS in a whole class setting both in England and internationally."
    • Relevant Quotes: 1) "The present study is a partial replication of that trial, and it represents the most rigorous test of the impact of PALS in a whole class setting both in England and internationally." (p. 3) 2) "Fuchs et al. (1997) explored effects in 12 active (and 20 control) schools in the US..." (p. 3) 3) "Sáenz et al. (2005) found significant effects on pre-/post- treatment improvement for comprehension only." (p. 3) Detailed Analysis: Criterion R requires that the specific study be independently replicated by a different research team in a different context and published in a peer-reviewed journal. The paper describes itself as a "partial replication" of an earlier PALS-UK trial, which is not evidence that this specific study has itself been replicated by an independent team. The paper also cites prior PALS studies (e.g., Fuchs et al., 1997; Sáenz et al., 2005), which are different contexts and are not presented as replications of this specific England PALS-UK 2022/23 school-randomized trial. Internet searching (as of 2026-04-14) did not identify a separate, peer-reviewed, independent replication study that explicitly reproduces this specific 2022/23 PALS-UK trial and reports results. Final Summary: Criterion R is not met because an independent replication of this specific study was not found.
  • A

    All-subject Exams

    • The study measures standardized reading outcomes only and does not assess all main subjects.
    • "We examine the treatment effect of PALS-UK on six outcomes to address RQ1 and RQ2. These include the primary outcome – reading attainment..."
    • Relevant Quotes: 1) "We examine the treatment effect of PALS-UK on six outcomes to address RQ1 and RQ2. These include the primary outcome – reading attainment, and secondary outcomes – reading skills (comprising reading comprehension and two measures of oral reading fluency), and affective factors (reading self-efficacy and motivation to read)." (p. 11) 2) "New PiRA is a series of termly UK curriculum-aligned standardized norm-referenced assessments." (p. 8) Detailed Analysis: Criterion A requires standardized exam-based assessment across all main subjects taught at that level (to detect curriculum trade-offs outside the intervention subject). The outcomes described in the paper are reading attainment, reading comprehension/fluency, and reading-related affective factors. No standardized assessments in other core subjects (e.g., mathematics) are described. Because Criterion E is met, Criterion A can be evaluated, but it still fails due to single-subject (reading-only) outcome coverage. Final Summary: Criterion A is not met because the study measures standardized outcomes only in reading, not across all main subjects.
  • G

    Graduation Tracking

    • The paper reports outcomes through the end of Year 5 only and does not report tracking to the end of primary school graduation or later milestones.
    • "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022) and again at endline after pupils in the treatment group had finished the program (end of Year 5, Summer 2023)."
    • Relevant Quotes: 1) "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022) and again at endline after pupils in the treatment group had finished the program (end of Year 5, Summer 2023)." (p. 8) 2) "Schools were asked to deliver PALS-UK three times a week for 20 school weeks, from October 2022 to May/June 2023." (p. 7) Detailed Analysis: Criterion G requires follow-up tracking through to graduation from the relevant educational stage (here, end of primary school / Key Stage 2 completion). The paper reports baseline at end of Year 4 and endline at end of Year 5 (Summer 2023). It does not report outcome data from the end of primary school (e.g., Year 6 / Key Stage 2 statutory assessments) and does not describe a design that follows pupils through that graduation milestone within this paper. Additional searching for follow-up peer-reviewed publications by the same author team reporting end-of-primary-school (graduation) outcomes for this specific 2022/23 trial cohort did not identify such a publication as of 2026-04-14. Final Summary: Criterion G is not met because the study does not provide evidence of tracking participants through to graduation.
  • P

    Pre-Registered

    • The protocol was preregistered before randomization, but the OSF preregistration date appears to be after baseline data collection had already occurred.
    • "The study was funded by the Education Endowment Foundation Accelerator Fund and the protocol was preregistered prior to randomization (https://educationendowmentfoundation.org.uk/projects- and-evaluation/projects/pals-uk-accelerator-fund and https://archive.org/details/osf-registrations-f2wgj-v1)."
    • Relevant Quotes: 1) "The study was funded by the Education Endowment Foundation Accelerator Fund and the protocol was preregistered prior to randomization (https://educationendowmentfoundation.org.uk/projects- and-evaluation/projects/pals-uk-accelerator-fund and https://archive.org/details/osf-registrations-f2wgj-v1)." (p. 5) 2) "The primary outcome (reading attainment) was assessed at baseline before randomization (end of Year 4, Summer 2022)..." (p. 8) 3) "Publication date 2022-08-15" (OSF archive record for osf-registrations-f2wgj-v1) 4) "By the end of Summer Term 2022, evaluators were informed of the identities of each participating school and the pupils within them. Baseline reading assessments were collected..." (EEF Evaluation Report for this trial, p. 23) Detailed Analysis: Criterion P requires that the protocol be preregistered before the study begins (i.e., before data collection starts), not merely before randomization. The paper states the protocol was preregistered prior to randomization and provides a public link to the archived OSF preregistration. However, the OSF archive record for the preregistration indicates a "Publication date 2022-08-15." The paper and the EEF Evaluation Report describe baseline assessment occurring by the end of Summer Term 2022, before randomization. Therefore, while preregistration occurred before randomization, the available evidence suggests baseline data collection had already occurred prior to the preregistration publication date. Final Summary: Criterion P is not met because the preregistration appears to have been published after baseline data collection began.