Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index; none of the 44 sources its claims cite is checked for support yet.

Language choice and students’ performance in mathematics and science: Evidence from Pakistan’s multilingual context

Open access
Bhutta, Sadia Muzaffar, Ansari, Aisha Naz, Ahmad, Sohail, Ali, Barkat·2026·American Journal of STEM Education

Abstract

In Pakistan’s post‑colonial multilingual context, language in education remains sensitive and complex. Most learners use their mother tongue at home, Urdu functions as the national lingua franca, and English, an official language, serves widely as the Language of Instruction (LoI) in private schools. These overlapping roles shape classroom interaction and student performance. This paper examines how language practices in Pakistani schools, particularly consistency across textbooks, instruction, and assessment, affect achievement in science and mathematics. Findings draw on a nationally representative sample of 15,391 students and 589 teachers from 153 elementary schools, using student achievement tests and classroom observations. Students exposed to consistent language use performed significantly better than those experiencing inconsistent language use in both subjects (p<0.001). The results show that language consistency, alongside school systems and teaching quality, influences performance, with effects strongest in classrooms with relatively better teaching quality.

Overview

CoChat Summary

TL;DR

In Pakistan’s multilingual elementary schools, students whose textbooks, classroom instruction, and assessments used the same language performed significantly better in mathematics and science than students exposed to inconsistent language use. The study reports significant main and interaction effects for language consistency with school system and teaching practices, though the t-test differences were small in magnitude.

Key findings

  • Students with consistent language use outperformed those with inconsistent language use in both subjects (p<0.001): mathematics consistent M=28.1 (SD=15.4) vs inconsistent M=26.5 (SD=15.5), t=-5.493, effect size r=0.05; science consistent M=34.9 (SD=16.1) vs inconsistent M=33.6 (SD=16.2), t=-4.431, effect size r=0.04.
  • Language consistency and school system had significant main effects on performance in mathematics and science (p<0.001 for all), with a significant interaction accounting for 7% of variance in mathematics and 12% in science. Private-school students with consistent language use scored higher in mathematics (M=36.4, SD=17.1) than inconsistent counterparts (M=30.1, SD=16.1), and in science private-school students with consistency scored M=45.8 (SD=16.9) vs government-school students with consistency M=38.4 (SD=17.1).
  • Language consistency and teaching quality also had significant main effects (p<0.001 for all), with a significant interaction accounting for 2% of variance in both subjects. In mathematics, students with consistent language and moderate teaching practices scored M=35.7 (SD=16.2) vs inconsistent counterparts M=30.9 (SD=16.3); in science, the corresponding scores were M=44.2 (SD=15.8) vs M=36.7 (SD=18.4).
  • Most private-school students (75%) experienced inconsistent language use, while 40% of government-school students encountered inconsistency across textbooks, classroom instruction, and assessment.

Methods

  • Cross-sectional survey drawn from a nationwide study in 153 schools (78 public, 75 private) across six regions of Pakistan, selected via multistage cluster sampling from 25 districts; all pupils in grades 5, 6, and 8 were invited, yielding 15,391 students, and 589 science and mathematics teachers were observed.
  • Instruments included a 3-point Classroom Observation Scale (COS) used to observe one lesson per teacher, grade-specific Science Achievement Tests (SATs) and Mathematics Achievement Tests (MATs) available in English, Urdu, and Sindhi, and demographic questionnaires. Test reliability (Cronbach’s alpha) ranged from 0.60 to 0.78; COS reliability was 0.79 for science and 0.69 for mathematics.
  • Analysis used independent t-tests to compare mean percentage scores by language consistency, effect size r, and two-way ANOVA to test interactions between language consistency and school system, and between language consistency and teaching practices. Teaching practices were benchmarked as weak, moderate, or good, but only weak and moderate categories appeared in the analysis.

Limitations

  • The paper does not state a formal limitations section.
  • Evident from the text: the cross-sectional design cannot establish causality; the authors call for future experimental studies to assess whether consistent language use makes a difference in students’ learning.
  • Evident from the text: t-test differences had small effect sizes (r=0.05 in mathematics, r=0.04 in science), and the teaching-practices interaction accounted for only 2% of variance in both subjects.
  • Evident from the text: no teachers reached the “good” teaching-practices benchmark, so the teaching-quality analysis is limited to weak and moderate practices.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.32674/pgf3k964

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Interaction between Consistent Use of Language and the System of Schools in Mathematics
Figure 1 Interaction between Consistent Use of Language and the System of Schools in MathematicsCC-BY · From the PDF
Figure 2 Interaction between Consistent Use of Language and System of Schools in Science
Figure 2 Interaction between Consistent Use of Language and System of Schools in ScienceCC-BY · From the PDF
Figure 3 Interaction between Consistent Use of Language and Teaching Practices in Mathematics
Figure 3 Interaction between Consistent Use of Language and Teaching Practices in MathematicsCC-BY · From the PDF
Figure 4 Interaction between Consistent Use of Language and Teaching Practices in Science
Figure 4 Interaction between Consistent Use of Language and Teaching Practices in ScienceCC-BY · From the PDF

Verification

Claims the paper makes

0 Valid · 1 Not valid · 24 Not checked · 59 Own results and interpretations

36 more claims

Create a free account to see all 84 claims and to verify the 17 references still open.

See all claims — free account
  • It may state two separate claims as one. 25
  • Probably not a substantial claim. 21

Create a free account to see these claims.

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.

5 of 9 · the rest aren’t loaded

4 more findings

Create a free account to see every finding of this check.

See all findings — free account

Citations

Cites

Show all 40 references

Research works printed but not linked: 9

  1. Benson, C., & Kosonen, K. (2013). A critical comparison of language-in-education policy and practice in four Southeast Asian countries and Ethiopia. In Multilingual Education and Sustainable Diversity Work (pp. 111–137). Routledge.
  2. Coleman, H. (2011). Developing countries and the English language. In H. Coleman (Ed.), Dreams and realities: Developing countries and the English language (pp. 1–14). British Council.
  3. Fareed, M., Ashraf, A., & Mushtaque, S. (2019). Medium of instruction in education: perceptions of teachers and students from Pakistani school, college and university. FWU Journal of Social Sciences, 13(3), 134–143.
  4. García, O., & Baker, C. (Eds.). (2007). Bilingual education: An introductory reader (Vol. 61). Multilingual matters.
  5. Mohanty, A. (2011). MLE and the double divide in multilingual societies: Comparing policy and practice in India and Ethiopia. In K. Kosonen & C. Benson (Eds.), Language issues in comparative education (pp. 138–150). Sense Publishers.
  6. Mustafa, Z. (2021). Some observations on language. In Reforming School Education in Pakistan & the Language Dilemma (pp. 42–50). Paramount Books (Pvt.) Ltd.
  7. Mustafaqulova, H. A. (2022). Teaching culture in the ESL classroom. Science and Education, 3(2), 817–820.
  8. Prah, K. K., & Brock-Utne, B. (2009). Introduction: Multilingualism – an African Advantage. In K. K. Prah & B. Brock-Utne (Eds.), Multilingualism – an African Advantage. A paradigm shift in African languages of instruction policies (pp. 1–17). CASAS.
  9. Wedikkarage, L. (2009). Science Education and English Medium: The Sri Lankan Experience. In B. Brock-Utne & G. Garbo (Eds.), Language is Power: The implications of Language for Peace and Development. Dar as Salaam.

Cited by

See all 3

Metadata

Authors
Bhutta, S. M., Ansari, A. N., Ahmad, S., & Ali, B.
Affiliations
  • Aga Khan University (PK)
  • Durham University (GB)
  • University of Cambridge (GB)
  • Karakoram International University (PK)
Venue
American Journal of STEM Education
Published
2026-05-26
Volume
22
Pages
289-316
ISSN
3065-1190 (ISSN-L)
OpenAlex
W7162869869
Cited by
3
References
40
Citations by year
2026: 2
FWCI
16.61
Type
journal article
Language
en
Open access
Locations
American Journal of STEM Education (publishedVersion, cc-by)
Topics
  • Social Sciences › Linguistics and Language › Multilingual Education and Policy
  • Social Sciences › Linguistics and Language › Second Language Learning and Teaching
  • Social Sciences › Education › School Choice and Performance
Keywords
language of instruction, language choice in education, language consistency, mathematics achievement, science achievement, multilingual education, Urdu language, English-medium instruction, private schools, teaching quality, elementary school, Pakistan

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 2026-11-16, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-16 (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: Language choice and students’ performance in mathematics and science: Evidence from Pakistan’s multilingual context. https://app.cochat.ai/papers/10.32674/pgf3k964
The paper
Bhutta, S. M., Ansari, A. N., Ahmad, S., & Ali, B. (2026). Language choice and students’ performance in mathematics and science: Evidence from Pakistan’s multilingual context. American Journal of STEM Education. https://doi.org/10.32674/pgf3k964

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.