Learning, under fire

Home · Claims · C25

C25 Test anxiety and results

alive Attack 1 on C25, 2026-11-03: survived (Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021) 1 attack: 1 survived · serves Q1 · last change 3 November 2026

Pupils and students with higher test anxiety get lower test scores and grades than calmer peers, at every school age studied: the link is negative and about r = −0.2 (small to moderate), not close to zero.

Go deeper

Hardest hit so far · attack #1, survivedOutside Robson et al.'s primary-school sample, the test-anxiety link may be weaker than about r = −0.1, or vanish in older students and adults.C26 · Stress at the test · relatedStress right before a memory test lowers recall about 0.2 SD, but only in studies that left out pill users.0 of 1 attack left it unchanged

Where it holds

All limits (4)

Inside the circle: the core of the claim. Outside: every bound and carve-out in the claim file, numbered, with the attack that forced or added it where the record shows one.

forced by a narrowing attackin the original wordingwritten at creation, or added without a narrowing

Warm cards in the list were added by later attacks; plain cards were in the original draft or written at creation.

Where C25 holdsThe core of C25 sits inside the circle. 4 bounds and carve-outs sit outside the boundary, numbered; the numbered list after the diagram gives each one in full with the attack that forced it.C25 · coreMore test anxiety goeswith lower scores andgrades, about r = −0.2.1An association, not proof thatanxiety causes lower results …at creation2Evidence on file: Robson,Johnstone, Putwain & Howard …at creation3Kill-condition test survived2026-11-03 (nightly attack): …#1 Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 20214Maths only, correlational, andpartly the same studies as von …#1 Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021

  1. An association, not proof that anxiety causes lower results (weaker pupils may also become more anxious).

    Written when the claim was created (29 October 2026); not forced by an attack

  2. Evidence on file: Robson, Johnstone, Putwain & Howard 2023 (Journal of School Psychology, DOI 10.1016/j.jsp.2023.02.003; full text): 76 studies, N = 53,617 children aged 5–12; maths r = −.21 (95% CI −.30 to −.13), literacy r = −.20 (−.29 to −.10), general school achievement r = −.23 (−.29 to −.16); some publication bias and heterogeneity. von der Embse, Jester, Roy & Post 2018 (Journal of Affective Disorders, DOI 10.1016/j.jad.2017.11.048; abstract only): 238 studies; significant negative relations to standardised tests, entrance exams and GPA, small to moderate, strongest in the middle grades.

    Written when the claim was created (29 October 2026); not forced by an attack

  3. Kill-condition test survived 2026-11-03 (nightly attack): Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021 (Educational Psychology Review 34(1), 363–399, DOI 10.1007/s10648-021-09618-5; open access, full text read): meta-analysis of academic anxiety and maths performance (177 studies, 906,311 participants overall); test anxiety and maths performance r = −.23 (95% CI −.26 to −.19), high heterogeneity, children versus adults not a significant moderator.

    Added with attack #1 (survived; it did not narrow the claim) · Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021 · 3 November 2026

  4. Maths only, correlational, and partly the same studies as von der Embse et al. 2018.

    Added with attack #1 (survived; it did not narrow the claim) · Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021 · 3 November 2026

0 of 4 bounds and carve-outs were forced by a specific attack. 1 attack hit the boundary and did not move it: #1.

How it changed

The ledger records 1 attack, but no earlier wording of C25 is on record, so the text cannot be replayed step by step. Only the current wording is shown.

Pupils and students with higher test anxiety get lower test scores and grades than calmer peers, at every school age studied: the link is negative and about r = −0.2 (small to moderate), not close to zero.

What would change this

What would kill it: A large meta-analysis finds no reliable negative link between test anxiety and test scores or grades, or a link weaker than about r = −0.1. Added as a claim on 29 October 2026, promoted from a candidate belief in the evidence library.

Attack record

1 attack: 1 survived. Counts of attacks, not a score.

Full attack list (1)
  1. #1 survived 3 November 2026 · Caviola, Toffalini, Giofrè, Mercader Ruiz, Szűcs & Mammarella 2021 · target: kill-condition

    Outside Robson et al.'s primary-school sample, the test-anxiety link may be weaker than about r = −0.1, or vanish in older students and adults.

    Strongest reading and verdict reasoning

    Steelmans to: The strongest version: a large meta-analysis across children and adults, separating test anxiety from maths anxiety, could show the test-anxiety link to be near zero once other anxiety forms are split off.

    Why: Nightly-attacks run of 2026-11-03 Europe/Stockholm. Test anxiety and maths performance r = −.23 (95% CI −.26 to −.19), high heterogeneity; children versus adults not a significant moderator. The link is reliably negative and well beyond −0.1. Limits: maths only (not grades or literacy), correlational, and partly overlapping studies with von der Embse et al. 2018.

Evidence

4 linked sources: 4 supports (first-pass draft, pending human review). supports: Robson et al. 2023, von der Embse et al. 2018, Hembree 1988, Caviola et al. 2021

Full claim record in the evidence library →