The DOI matches its Crossref record; 88 of the 90 references it cites were found in a registry or an index; none of the 30 sources its claims cite is checked for support yet.
Mathematics Interventions for Children with Special Educational Needs: A Meta-Analysis
Open accessAbstract
This article presents the results of a meta-analysis of 58 studies of mathematics interventions for elementary students with special needs. Interventions in three different domains were selected: preparatory mathematics, basic skills, and problem-solving strategies. The majority of the included studies described interventions in the domain of basic skills. In general, these interventions were also the most effective. Furthermore, a few specific characteristics were found to influence the outcomes of the studies. In addition to the duration of the intervention, the particular method of intervention proved important: Direct instruction and self-instruction were found to be more effective than mediated instruction. Interventions involving the use of computer-assisted instruction and peer tutoring showed smaller effects than interventions not including these supports.
Overview
CoChat Summary
TL;DR
This meta-analysis of 58 studies (61 effect sizes; 2,509 children) examined mathematics interventions for elementary students with special educational needs across preparatory arithmetic, basic facts, and problem solving. Basic-skills interventions were most studied and generally most effective, while self-instruction and direct instruction outperformed mediated/assisted instruction, and computer-assisted instruction and peer tutoring showed smaller effects.
Key findings
- Across 58 studies, 13 interventions targeted preparatory arithmetic, 31 targeted basic facts, and 17 targeted problem solving; basic facts were investigated most, χ²(2, N = 61) = 7.897, p = .019. Weighted effect sizes were 1.14 for basic facts, 0.92 for preparatory skills, and 0.63 for problem solving.
- Four variables explained 69% of between-studies variance: design (γ = 1.4, p = .000; single-subject studies had higher effects than group studies), duration (γ = -0.01, p = .032; longer interventions had less effect), domain (problem solving was less effective than preparatory and basic skills), and method (self-instruction was more effective than direct or mediated/assisted instruction; self-instruction vs. mediated/assisted γ = 0.66, p = .009).
- Weighted effect sizes by method were 1.45 for self-instruction, 0.91 for direct instruction, and 0.34 for mediated/assisted instruction. Computer-assisted instruction had a weighted effect of 0.51 versus 1.05 for teacher-delivered instruction; peer tutoring had a weighted effect of 0.87 versus 0.96 without peer tutoring.
- In domain-specific multilevel analyses, basic-facts interventions were explained by control condition, peer tutoring, mean age, and method; problem-solving interventions were explained by control condition, year, special needs, peer tutoring, and CAI, with peer tutoring and CAI being less effective.
Methods
- Meta-analysis of empirical studies published 1985–2000, identified through Current Contents, ERIC, PsychLit, and SSCI, plus hand searches and reference checks; 656 initial references yielded 220 retained publications and finally 58 studies, producing 61 effect sizes.
- Inclusion criteria: elementary/kindergarten mathematics intervention, mean participant age not higher than 12, children with mathematical difficulties, between-subjects or within-subjects control condition, at least three children, and sufficient quantitative information to calculate effect sizes.
- Effect sizes were calculated as Cohen’s d, with a maximum effect size of 3; analyses used a random effects model and multilevel regression analysis of variance via VKHLM. Coding covered study characteristics, sample characteristics, and treatment parameters/components.
Limitations
- Stated in the paper: including only published journal articles may weaken external validity because of a tendency to publish only studies with significant positive effects; only English-language articles were included; conclusions are based only on studies published in well-known journals.
- Stated in the paper: single-subject and group-design effect sizes were calculated differently, and single-subject studies were too few and too homogeneous to analyze separately; some high effect sizes were capped at 3; some participant ages were estimated.
- Evident from the text: the preparatory-arithmetic separate meta-analysis had low numbers for CAI, peer tutoring, self-instruction, and mediated instruction, so those variables were not analyzed; the authors note that separate domain analyses had small Ns and many variables.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 0 Not valid · 18 Not checked · 98 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A total of 2,509 children with special mathematics needs was studied.”
plainn = 2,509- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Also, an effect of the special needs of the students was found, because the interventions for students with learning disabilities showed higher effect sizes than the interventions for the other groups; 23 studies described interventions for lowperforming/at-risk children, 23 for children with learning disabilities, 8 for children with mild or educable mental”
plainstates an associationn = 23- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, no significant differences were found between the effect sizes of the three different domains.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The domain of basic facts has been investigated most, χ 2 (2, N = 61) = 7.897, p = .019.”
plainn = 61 ·- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Interventions involving the use of computerassisted instruction and peer tutoring showed smaller effects than interventions not including these supports.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
65 more claims
Create a free account to see all 116 claims and every source verdict.
- Probably not a substantial claim. 23
- It may state two separate claims as one. 29
- The quote only partly states this claim. 2
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Interventions for children with special needs showed increases in effect size with increasing age, suggesting a potentially later sensitive period (up to age 12, in a cutoff of a meta-analysis; Kroesbergen and Van Luit, 2003)”
From Does puberty mark a transition in sensitive periods for plasticity in the associative neocortex? (2017) · cites this paper as Kroesbergen and Van Luit, 2003
Statistics that don’t add up
Each reported p-value recomputed from its test statistic and degrees of freedom.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“F (1, 16) = 2.71, p < .05”
Reported p < .05 · recomputed p = .119
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“F (1, 31) = 1.8, p = .005”
Reported p = .005 · recomputed p = .189
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
“F (1, 20) = 2.72, p < .05”
Reported p < .05 · recomputed p = .115
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“t (15) = 1.86, p < .05”
Reported p < .05 · recomputed p = .083
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“F (1, 12) = 3.27, p = .05”
Reported p = .05 · recomputed p = .096
Statistics recomputed
Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.
- Doesn’t add upF(1, 31) = 1.8reported p = .005 · recomputed p = .189The two fall on opposite sides of .05
“F (1, 31) = 1.8, p = .005”
- Doesn’t add upF(1, 20) = 2.72reported p < .05 · recomputed p = .115The two fall on opposite sides of .05
“F (1, 20) = 2.72, p < .05”
- Doesn’t add upF(1, 16) = 2.71reported p < .05 · recomputed p = .119The two fall on opposite sides of .05
“F (1, 16) = 2.71, p < .05”
- Doesn’t add upF(1, 12) = 3.27reported p = .05 · recomputed p = .096
“F (1, 12) = 3.27, p = .05”
- Doesn’t add upt(15) = 1.86reported p < .05 · recomputed p = .083
“t (15) = 1.86, p < .05”
14 more recomputed tests
Create a free account to see every test the Stats check recomputed.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Question 2: Effects of Single Variables · p. 6
“Studies in which the control condition received a different intervention ( n = 30), showed higher effect sizes than studies without a different intervention in the control condition ( n = 31).”
§ Question 2: Effects of Single Variables · p. 14
“| no intervention | 30 | 701 | 1.87 | 1.01 | 1.51 | | intervention | 31 | 1,808 | 0.54 | 0.59 | 0.51 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Question 2: Effects of Single Variables · p. 6
“The average number of participants in the studies was 41.1, with a range of 3 to 136.”
§ Question 2: Effects of Single Variables · p. 10
“| Mevarech (1985) | 7 | 204 | At risk |”
Citations
Cites
National Council of Teachers of Mathematics
(1979) · The Mathematics Teacher · doi:10.5951/mt.72.8.0637 · Cited by 8,499
The Handbook of Research Synthesis.
Howel, D., Cooper, H., & Hedges, L. V. (1995) · Biometrics · doi:10.2307/2533295 · Cited by 3,642
Parametric measures of effect size.
Rosenthal, R. (1994) · Cited by 1,847
Hox, J. J. (1995) · Utrecht University Repository (Utrecht University) · Cited by 1,168
Summarizing Single-Subject Research: Issues and Applications
Scruggs, T. E., & Mastropieri, M. A. (1998) · Behavior Modification · doi:10.1177/01454455980223001 · Cited by 706
Children's mathematical development: Research and practical applications.
Geary, D. C. (1994) · doi:10.1037/10163-000 · Cited by 565
(1994) · Cited by 463
Swanson, H. L., & Hoskyn, M. (1998) · Review of Educational Research · doi:10.3102/00346543068003277 · Cited by 409
Comparing effect sizes of independent studies.
Rosenthal, R., & Rubin, D. B. (1982) · Psychological Bulletin · doi:10.1037//0033-2909.92.2.500 · Cited by 406
Peer-model attributes and children's achievement behaviors.
Schunk, D. H., Hanson, A. R., & Cox, P. D. (1987) · Journal of Educational Psychology · doi:10.1037/0022-0663.79.1.54 · Cited by 363
Research works printed but not linked: 19
- References marked with an asterisk indicate studies included in the meta-analysis
- Beirne-Smith, M., 1991, Exceptional Children, 57, 330
- Bryk, A.S., 1994, HLM 2/3. Hierarchical linear modeling with the HLM/2L and HLM/3L programs
- Harris, C.A., 1995, Learning Disabilities Research & Practice, 10, 180
- Hasselbring, T.S., 1996, Journal of Computing in Childhood Education, 7, 199
- Hativa, N. (1994). Cognitive and affective effects of computer-based arithmetic practice on the lowest achieving students. In J. E. H. Van Luit (Ed.), Research on learning and instruction in kindergarten and primary school (pp. 303—327). Doetinchem, The Netherlands: Graviant.
- Hox, J.J., 1997, Tijdschrift Voor Onderwijs Research, 22, 16
- Koscinski, S.T., 1993, Learning Disabilities Research & Practice, 8, 157
- Kroesbergen, E.H., 2002, Mathematics instruction for low-achieving students: An intervention study
- Malabonga, V., 1995, Child Study Journal, 25, 79
- Marsh, L.G., 1996, Learning Disabilities Research & Practice, 11, 58
- Mastropieri, M.A., 1991, Education and Training in Mental Retardation, 26, 115
- Mercer, C.D., 1997, Students with learning disabilities
- Miller, S.P., 1998, Exceptional Children, 31, 24
- Miller, S.P., 1993, Learning Disabilities Research & Practice, 8, 89
- National Council of Teachers of Mathematics., 2000, Principles and standards for school mathematics
- Perry, P., 1992, Education and Training in Mental Retardation, 27, 273
- Podell, D.M., 1992, Education and Training in Mental Retardation, 27, 200
- Schopman, E.A.M., 1996, Journal of Cognitive Education, 5, 117
Cited by
Mathematics and Learning Disabilities
Geary, D. C. (2004) · Journal of Learning Disabilities · doi:10.1177/00222194040370010201 · Cited by 1,245
Gersten, R., Chard, D. J., Jayanthi, M., Baker, S. K., Morphy, P., & Flojo, J. (2009) · Review of Educational Research · doi:10.3102/0034654309334431 · Cited by 689
Effective Programs in Elementary Mathematics: A Best-Evidence Synthesis
Slavin, R. E., & Lake, C. (2008) · Review of Educational Research · doi:10.3102/0034654308317473 · Cited by 407
Computer-assisted intervention for children with low numeracy skills
Räsänen, P., Salminen, J., Wilson, A. J., Aunio, P., & Dehaene, S. (2009) · Cognitive Development · doi:10.1016/j.cogdev.2009.09.003 · Cited by 314
“In their meta-analyses Kroesbergen and van Luit (2003) and Malofeeva (2005) concluded that a computer was less effective than a teacher in assisting learning.”
BackgroundA Meta-Analysis on Teaching Mathematics to Students with Significant Cognitive Disabilities
Browder, D. M., Spooner, F., Ahlgrim-Delzell, L., Harris, A. A., & Wakemanxya, S. (2008) · Exceptional Children · doi:10.1177/001440290807400401 · Cited by 266
Metadata
- Authors
- Kroesbergen, E. H., & Van Luit, J. E. H.
- Affiliations
- Utrecht University (NL)
- Utrecht University, The Netherlands
- Venue
- Remedial and Special Education
- Publisher
- SAGE Publications
- Published
- 2003-03
- Volume
- 24
- Issue
- 2
- Pages
- 97-114
- ISSN
- 0741-9325 (ISSN-L)
- 1538-4756
- OpenAlex
- W2100113892
- Cited by
- 549
- References
- 90
- Citations by year
- 2026: 8 · 2025: 17 · 2024: 14 · 2023: 15 · 2022: 24 · 2021: 23 · 2020: 34 · 2019: 32 · 2018: 23 · 2017: 32 · 2016: 34 · 2015: 39 · 2014: 49 · 2013: 46 · 2012: 30
- FWCI
- 8.19
- Type
- journal article
- Language
- en
- Open access
- green
- hdl.handle.net
- Licence
- journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
- Locations
- Remedial and Special Education (publishedVersion)
- Radboud Repository (Radboud University) (submittedVersion)
- Topics
- Social Sciences › Psychology › Developmental and Educational Psychology › Cognitive and developmental aspects of mathematical skills
- Social Sciences › Psychology › Developmental and Educational Psychology › Behavioral and Psychological Studies
- Social Sciences › Education › Mathematics Education and Pedagogy
- Keywords
- mathematics intervention, special educational needs, elementary school students, basic mathematical skills, problem-solving strategies, direct instruction, self-education, mediated instruction, computer-assisted instruction, peer tutoring, intervention effectiveness
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Mathematics Interventions for Children with Special Educational Needs: A Meta-Analysis. https://app.cochat.ai/papers/10.1177/07419325030240020501
- The paper
- Kroesbergen, E. H., & Van Luit, J. E. H. (2003). Mathematics Interventions for Children with Special Educational Needs: A Meta-Analysis. Remedial and Special Education. https://doi.org/10.1177/07419325030240020501
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.