SAT scores underpredict the academic performance of high school girls, who average lower scores than boys despite higher grades¶

Claim · All claims
Evidence · 1 study · theoretical r1 · q2
Currency · newest study 1989 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 1992: tool and model accuracy changes quickly
Evidence quality · No study behind this claim is a peer-reviewed article with a DOI: it rests on an ERIC document (a report, conference paper or thesis). It is kept because 3 design pages cite it; treat it as a lead to check, not an established finding.

Subclaims¶

q2 i? Senior high school girls averaged 61 points lower than boys on the SAT (50 points lower in math, 11 in verbal) despite having higher grades in all high school subjects and in college, indicating systematic underprediction of female performance. → Sex and Race Differences on Standardized Tests: Oversight Hearings before the Subcommittee on Civil and Constitutional Rights of the Committee on the Judiciary 1989

Evidence¶

Sex and Race Differences on Standardized Tests: Oversight Hearings before the Subcommittee on Civil and Constitutional Rights of the Committee on the Judiciary 1989¶

Sex and Race Differences on Standardized Tests: Oversight Hearings before the Subcommittee on Civil and Constitutional Rights of the Committee on the Judiciary, House of Representatives, One Hundredth Congress, First Session. (1989). https://eric.ed.gov/?id=ED312276

q2 · i? · the article prints no effect size for this finding · theoretical · r1

Testimony of Phyllis Rosser, a consultant on sex bias in testing, at the April 23, 1987 hearing, reporting College Board score data. She states girls averaged "61 points lower than boys on the SAT" while earning higher grades, and argues that if the test accurately predicted freshman grades girls would score 20 points higher than boys.

"Although females have higher grades than males in all subjects in high school and higher college grades, even the freshman year, senior high school girls averaged 61 points lower than boys on the SAT last year, 50 points lower on the math section, and 11 points lower on the verbal."

Studies bearing on this claim¶

Does a standardized score predict later performance equally well for each group of learners it is used to select? This table gathers the studies behind this claim and the 5 narrower claims that bear on it: 5 distinct studies (4 associational, 1 theoretical), q2–q3, 0 of 5 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Learner Variables¶

• Gender — predictor: outcomes differ by group: a gap reflecting circumstances and treatment, not a cause

Related Claims¶

• Systematic racial, ethnic and gender differences in grades persist among students with the same test scores, attendance, courses and teachers — the general claim this finding bears on, with the studies behind it

• Traditional gifted identification tests cannot fully determine learning potential and omit classroom-learning components — the general claim this finding bears on, with the studies behind it

• Congressional hearing testimony reports, citing a 1979 ETS study second-hand, that females tend to do better on test items with female or neutral figures than on items with male figures — related
• Testimony at a 1987 congressional hearing reported 1985 SAT averages in which Black women scored 43 points below Black men and 264 points below white men, which the witness called a double penalty for minority females — related
• Women admitted to MIT with lower SAT math scores perform as well as men in freshman math classes — a narrower finding that bears on this claim
• PSAT-based National Merit scholarship selection distributes awards disproportionately to boys, with girls receiving 36 percent of scholarships — related
• Testimony at a 1987 congressional hearing reported that the Johns Hopkins talent search, which admitted by SAT score, invited over 2,500 boys but 1,081 girls in one summer although equal numbers took the test — related
• Differences in grading practices across teachers and schools are small relative to students' actual academic preparation, and GPAs average out those differences — related
• Ninth-grade GPAs varied substantially by gender, race/ethnicity, neighborhood SES, incoming test scores, course level, and school — related
• Girls score lower than boys on timed standardized math and science tests at the same course level, and SAT math scores underpredict women's college math performance — related
• Prospective regression analysis of the same data found somewhat larger sex differences in the same direction — related
• The score gaps constitute evidence of sex differences in the validity of the SAT-M as a predictor of college math performance, though not of their cause — a narrower finding that bears on this claim
• Women scored about 33 SAT-M points lower than men who took the same math course and received the same grade — related
• The fraction of women enrolling in mathematics degree programs in the UK and the Netherlands has remained constant despite rising female participation in higher education — related
• Boys received more process praise and scored higher in 4th grade math than girls in this sample — related
• Girls report higher school engagement at time 1 and higher cumulative GPA than boys — related
• Girls were on average more accurate than boys in equation rearrangement, with no significant difference in speed — related
• Teacher-rated grades show girls rated slightly higher than boys, but gender is unrelated to test performance — related
• Gender differences in global performance judgments follow domain stereotypes: men higher in mathematics, women higher in reading, with no gender differences in bias — related
• The learning benefit of mathematics test practice is larger for boys than for girls — related
• Single-sex math classrooms raised female students' math performance by 7–10% over co-educational programs — related
• Late-arriving English Learners had cumulative GPAs similar to the district average but substantially lower SAT scores — related
• Girls scored lower than boys on fifth grade Numerical Operations, with the gap smaller in earlier grades, but no sex difference in fifth grade word reading — related
• Women were less accurate than men on the opposites tests but were less mature and less advanced in scholarship — related
• In this group the average girl's grades equal or exceed the average boy's in all six subjects, with the largest advantage in Latin — related