Showing 84 database articles primarily about Test Scores

Currently filtering by:

For whatever topics you select (currently: Test Scores):

Include articles that mention this topic 136

Only articles primarily about this topic 84

The Elite Illusion: Achievement Effects At Boston And New York Exam Schools

Atila Abdulkadiroglu, Joshua Angrist and Parag Pathak Econometrica

Date Posted:: May 22, 2020
Is Database:: Database

Attending elite exam schools in Boston & New York yields only scattered academic gains, with most students likely achieving similar success without attending.

The study reveals that attending elite exam schools in Boston and New York yields only scattered academic gains, with the most notable improvement being in 10th grade English scores, particularly for minority students. Despite the high peer achievement levels at these schools, the data suggests that peer effects are not significant in enhancing educational outcomes. Most students would likely have achieved similar success without attending these schools, indicating that the perceived benefits may be overstated. While exam school graduates might earn higher wages, these gains are likely due to factors other than increased cognitive achievement or peer composition. The findings challenge the assumption that group averages strongly influence individual outcomes, highlighting the risk of specification error in peer analysis. Overall, the results suggest that the elite status of these schools may not translate into substantial academic advantages for their students.

Atila Abdulkadiroglu, Joshua Angrist and Parag Pathak, “The Elite Illusion: Achievement Effects At Boston And New York Exam Schools,” Econometrica, January 2014, https://economics.mit.edu/files/9518

“…The results reported here suggest that an exam school education produces only scattered gains for applicants, even among students with baseline scores close to or above the mean in the target school. Because the exam school experience is associated with sharp increases in peer achievement, these results weigh against the importance of peer effects in the education production function. Our results also fail to uncover systematic evidence of racial composition effects. The outcome most strengthened by exam school attendance appears to be the 10th grade English score, a result driven partly by gains for minorities.Given the history of racial preferences (and their more recent elimination) in Boston’s exam schools, this finding seems worth further exploration. Overall, however,while the exam school students in our samples typically have good outcomes, most of these students would likely have done well without the benefit of an exam school education. Of course, test scores and peer effects are only part of the exam school story. It may be that preparation for exam school entrance is itself worthwhile.The RD design captures the impact of peer composition and possibly other changes at admissions cutoffs, while ignoring effects common to applicants on both sides. Likewise, unique features of an exam school education may boost achievement in specific subject areas. Students who attend Boston Latin School almost certainly learn more Latin than they would have otherwise. The many clubs and activities found at some exam schools may expose students to ideas and concepts not easily captured by achievement tests or our post-secondary outcomes. It is also possible that exam school graduates earn higher wages…Still, the estimates reported here suggest that any labor market gains are likely to come through channels other than peer composition and increased cognitive achievement. Can these results be reconciled with those of other studies reporting a strong association between group averages and individual outcomes? Every context is different, and the absence of peer effects in one setting does not prove that such effects are unimportant elsewhere.At the same time, it is worth emphasizing the high risk of specification error in peer analysis….the relationship between an individual-level variable and any group average of this variable is essentially mechanical. Likewise, any regression of one outcome variable on the group average of another outcome variable is biased toward a finding of peer effects due to the presence of common shocks. Random assignment to groups fails to solve either of these problems. Finally, any regression of an individual outcome on a combination of observationally varying individual- and group-level treatments is almost sure to produce something that looks like a peer effect for reasons related to the information content of the average (see, e.g., Acemoglu and Angrist (2000) and Moffitt (2001)). It is perhaps unsurprising, therefore, that experimental or quasi-experimental manipulation of predetermined peercharacteristics is less likely to uncover a peer effect than are other study designs. Estimates from the Moving to Opportunity demonstration, for example, which manipulated peer composition by randomly assigning housing vouchers, are consistent with the results reported here in offering little evidence of causal peer effects…”

Not surprising to me that school doesn’t change cognitive outcomes (remember all these kids were motivated enough to take the entrance exam in the first place) but hard to believe it doesn’t have an impact.

Ed Comment:I’d also like to see his papers on the no elite effect and the one that says the last student admitted to Stuyvesant isn’t hurt by it, although the latter seems less interesting because it’s a we didn’t find any statistically significant effect rather than we found a statistically significant effect, which seems much more reliable to me. I think the benefit of being with elite peers is that the classes can move faster and reach further. When you are with non-elite peers, the instruction is held back. My kid’s schools splits the classes so the smart kids can move faster and dig deeper. If the amount of instruction matters as he claims is the benefit of no excuses characters then, it’s hard to see how this wouldn’t be true.

Steve Comment: The “Elite Illusion” By Atila (Duke) and Josh and Parag (MIT). Underwhelming to me, he isn’t capturing is the counterfactual, what happens to the laststudent admitted to Stuyvesant if he/she doesn’t go over the long term. The expectation at Stuy is to go to an elite school, because that gets you in the door at an institution that provides training (The Bain Consultings or JMP of the world) the expectation at other schools isn’t that. Note:“…he many clubs and activities found at some exam schools may expose students to ideas and concepts not easily captured by achievement tests or our post-secondary outcomes. It is also possible that exam school graduates earn higher wages….”

PISA 2025 Results (Volume I)

OECD Staff Organisation for Economic Co-operation and Development

Date Posted:: September 8, 2026
Is Database:: Database

Mean OECD math and reading scores fell roughly 25 PISA points—about a year of schooling—between 2018 and 2025, alongside reduced motivation and engagement and increased carelessness, exemplified by a near-doubling of “hasty readers.”

In math, performance remained close to the 2003 level up to 2018, then dropped sharply btw 2018 and 2025. Mean performance dropped by 9 score points in mathematics and about 14 score points in reading btw 2022 and 2025. Over the 2018-2025 period, the largest drops [in reading] were among the most advantaged 25% of socio-economic status, and the socio-economic gap with the most disadvantaged students reduced somewhat, by 11 points, on average across 35 OECD countries. In mathematics, the share of disadvantaged students scoring below proficiency Level 2 increased in previous cycles and remained high in 2025; among advantaged students, the share scoring below proficiency Level 2 increased, reaching 18% in 2025, about 4pp. Students’ reports of their motivation for learning and engagement with school also declined between 2022 and 2025. The proportion of “hasty readers” – those who gave fast and incorrect responses – increased by about five percentage points, on average, from almost 7% in 2018 to 11% in 2025. Analyses for math, also show an increase in the proportion of hasty responses between 2022 and 2025.

Takeaways by Macro Roundup^® AI

The share of “hasty readers” — students giving fast, incorrect responses — nearly doubled from 7% in 2018 to 11% in 2025, with math showing a parallel rise between 2022 and 2025, signaling deteriorating effort and engagement rather than pure skill loss.
Among the most advantaged quartile of students, the share scoring below math proficiency Level 2 reached 18% in 2025, up approximately 4 percentage points, indicating that academic decline is no longer concentrated among disadvantaged populations.

Standardized Test Scores and Academic Performance at a Public University System

Theodore Joyce, Mina Afrouzi Khosroshahi, Sarah Truelsch, Kerstin Gentsch, et al. National Bureau of Economic Research

Date Posted:: March 25, 2026
Is Database:: Database
Is Important:: Important

SAT scores are known to be better predictors of college and post-college success in Ivy+ universities. By contrast, this study of a public urban university system with 11 colleges shows that for the 2010–2019 cohort, high school GPA strongly dominates SAT scores.

Many of the country’s most selective colleges have returned to requiring the SAT, based on evidence that the test not only predicts student success, but facilitates the identification of qualified applicants from disadvantaged backgrounds. The results from our study of a large public urban university system find the opposite: high school grades are a vastly superior predictor of student academic success than is the SAT. We undertake an exercise with binned scatter plots of six year graduation rates against HSGPA [High School Grade Point Average] stratified by URM [under-represented minority] and whether the student received the maximum Pell grant (Figures 1a and 1b). [Note that SAT is included in the regression so the monotonic linear effect of GDP remains after controlling for SAT score]. We find that URM are less likely to graduate in six years at every level of HSGPA than are non-URM. In contrast, we find no similar difference when we stratify the plot by whether the student received the max Pell grant or not, a strong proxy for a low-income household (Figure 1b). The clear takeaway from the analysis of the 2010-2019 cohorts is the dominance of HSGPA in predicting student academic success. The results are in direct contrast to recent high-profile analyses of students from Ivy-plus colleges that demonstrated the superiority of SAT over HSGPA in predicting not only first-year academic outcomes, but post-college success as well.

The Lost Generation

Jacob Savage Compact Magazine

Date Posted:: December 17, 2025
Is Database:: Database
Is Important:: Important

Jacob Savage argues that the widespread discrimination against young white men in “high-status” sectors such as media and tech during the 2010s is a key driver of contemporary political dynamics among young men.

In 2011, the year I moved to Los Angeles, white men were 48% of lower-level TV writers; by 2024, they accounted for just 11.9%. The Atlantic’s editorial staff went from 53% male and 89% white in 2013 to 36% male and 66% white in 2024. White men fell from 39% of tenure-track positions in the humanities at Harvard in 2014 to 18% in 2023. The white men shut out of the culture industries didn’t surge into other high-status fields. At Google, white men went from nearly half the workforce in 2014 to less than a third by 2024—a 34% decline. In 2014, at Amazon, entry-level “professionals”—college graduates just starting out—were 42.3% white male. These were the employees who, if they’d advanced normally over the next decade, would be the mid-level managers of today. But mid-level Amazon managers fell from 55.8% white male in 2014 to just 33.8% in 2024—a decline of nearly 40%. In retrospect, 2014 was the hinge, the year DEI became institutionalized across American life.

Twelfth-Grade Math and Reading Scores in U.S. Hit New Low

Matt Barnum Wall Street Journal

Date Posted:: September 9, 2025
Is Database:: Database
Is Important:: Important

New National Assessment of Educational Progress results show the share of American high school seniors who scored basic or above on national tests fell to the lowest level on record: 68% in reading and 55% in math.

American high-school seniors’ scores on major math and reading tests fell to their lowest levels on record, according to results released Tuesday by the U.S. Education Department. Twelfth-graders’ average math score was the worst since the current test began in 2005, and reading was below any point since that assessment started in 1992. The share of 12th-graders who were proficient slid by 2 percentage points between 2019 and 2024—to 35% in reading and 22% in math. The results are from tests that are part of the National Assessment of Educational Progress, administered to tens of thousands of students in early 2024. In reading, two-thirds of seniors could determine the purpose of a persuasive essay, but only one in five was able to draw a conclusion from such an essay, supported by the text. In math, 60% of students deduced the population of an area using information on size and density, while just under half correctly turned a real-world scenario into an algebraic expression.

Standardized Test Scores and Academic Performance at Ivy-Plus Colleges

John Friedman, Bruce Sacerdote, Douglas Staiger, and Michele Tine National Bureau of Economic Research

Date Posted:: March 18, 2025
Is Database:: Database
Is Important:: Important

Students at Ivy-plus colleges whose SAT/ACT scores were at the 99th percentile had 1st-year college GPAs 0.43 points higher than those who had scored at the 75th percentile, while high school GPAs had little predictive power. @John_N_Friedman

We analyze admissions and transcript records for students at multiple Ivy-Plus colleges to study the relationship between standardized (SAT/ACT) test scores, high school GPA, and first-year college grades. Standardized test scores predict academic outcomes with a normalized slope four times greater than that from high school GPA, all conditional on students’ race, gender, and socioeconomic status. Standardized test scores do not underpredict college performance for students from less advantaged backgrounds. Collectively these results suggest that standardized test scores provide important information to measure applicants’ academic preparation that is not available elsewhere in the application file. While some of the raw correlation between these variables relates to students’ demographic characteristics, much remains even when controlling for these variables. In contrast, high school GPA does not predict academic performance nearly as well. We also find no evidence of bias against students from less advantaged backgrounds, as students from such backgrounds do not outperform (and in some cases underperform) other students with the same test score

Can Gifted Education Help Higher-Ability Boys from Disadvantaged Backgrounds?

David Card, Eric Chyn, and Laura Giuliano National Bureau of Economic Research

Date Posted:: February 21, 2025
Is Database:: Database

Card et al, find that the 25-30 percentage point college enrollment gap between disadvantaged boys and girls with IQs of ~116 closed among students who qualified for a “gifted” program. @EricChyn

We study the effects of being identified as gifted in elementary school on subsequent test scores, courses and grades, and on the probability of entering college, for disadvantaged children in a large urban school district (“the District”). The law specifies a threshold of 130 [IQ] points for non-disadvantaged students, but allows districts to select a lower “Plan B” threshold [116 pts] for FRL and ELL students. The strict IQ threshold for gifted status provides the basis for a regression discontinuity (RD) evaluation of the impacts of being identified as gifted. Figure 6 illustrates how these impacts affect gender gaps by presenting the estimated mean potential outcomes for the complying boys (left side) and complying girls (right side) using an 8-point bandwidth. The estimated college entry rate for male compliers who narrowly miss the gifted threshold is 46%; the 28 ppt treatment effect of passing the threshold raises this to 74%, which equals the rate for female compliers who narrowly miss the threshold - being identified as gifted raises the college entry rate of marginally eligible boys to the same rate as girls.

Attending elite exam schools in Boston & New York yields only scattered academic gains, with most students likely achieving similar success without attending.

Mean OECD math and reading scores fell roughly 25 PISA points—about a year of schooling—between 2018 and 2025, alongside reduced motivation and engagement and increased carelessness, exemplified by a near-doubling of “hasty readers.”

Are global education systems failing to teach core skills?

Takeaways by Macro Roundup® AI

Related Articles:

SAT scores are known to be better predictors of college and post-college success in Ivy+ universities. By contrast, this study of a public urban university system with 11 colleges shows that for the 2010–2019 cohort, high school GPA strongly dominates SAT scores.

Related Articles:

Jacob Savage argues that the widespread discrimination against young white men in “high-status” sectors such as media and tech during the 2010s is a key driver of contemporary political dynamics among young men.

Related Articles:

New National Assessment of Educational Progress results show the share of American high school seniors who scored basic or above on national tests fell to the lowest level on record: 68% in reading and 55% in math.

Related Articles:

Students at Ivy-plus colleges whose SAT/ACT scores were at the 99th percentile had 1st-year college GPAs 0.43 points higher than those who had scored at the 75th percentile, while high school GPAs had little predictive power. @John_N_Friedman

Related Articles:

Card et al, find that the 25-30 percentage point college enrollment gap between disadvantaged boys and girls with IQs of ~116 closed among students who qualified for a “gifted” program. @EricChyn

Related Articles:

Takeaways by Macro Roundup^® AI