E-bulletin, 2007 October 12

There is good news to report in Georgia education. Since 2003, the gap between Georgia and the national average on the Critical Reading and Math sections of the SAT has shrunk by 14 points.
In that time, Georgia's Critical Reading + Math scores have increased by 5 points, while the average across the country has dropped by 9.
In 2005-06, the Writing section was added to the SAT, increasing the possible total from 1600 points to 2400 points. In 2007, this meant that Georgia's overall average trailed the national average, 1472 to 1511.

Critical Reading + Math Scores: 2003 and 2007

Who's beating the national average?

System Rome City Fulton Oconee Decatur City Cherokee Fayette Forsyth Bleckley Jefferson City Floyd Cobb Gwinnett Marietta City Whitfield Columbia National Ave. Georgia Ave.

CR M W Total Test-takers Of the school systems in Georgia with 20

538 547 536 1621

131 or more 2007 SAT test-takers, 15 had total

527 540 528 1595 524 534 517 1575

3744 390 scores above the national average of 1511

537 527 510 1574

147 (Critical Reading +Math +Writing).

519 521 508 515 529 504 517 521 508 515 530 500 505 549 506

1548 1548 1546 1545 1560

1129 Five of these school systems ` scores 1480 include only one high school, but the 1065
43 remaining 10 had multiple schools and 81 still beat the national average.

516 515 504 515 517 502 504 527 493 505 519 498

1535 1534 1524 1522

153 All told, these 15 school systems 5221 represent approximately 36% of Georgia's 6155 255 2007 SAT test-takers.

512 505 498 1515

157 Unfortunately, this also means that the

505 518 490 1513

1029 average for every other school system in

502 515 494 1511 1,494,531 Georgia was below the national average.

494 495 483 1472

59,562

Georgia's 2007 SAT participation rate was 69% compared with the national average of 48%.

Do poverty rates predict SAT scores?
Yes, a school's poverty rate is a strong predictor of its standardized test scores. In fact, comparing schools and systems within Georgia based on their scores alone is probably not the clearest way to compare them.
To bypass this inherent test score advantage enjoyed by low-poverty schools, the Governor's Office of Student Achievement (GOSA) conducted an apples-to-apples comparison of schools' SAT success. It used a regression analysis that evaluated how well Georgia public schools' free/reduced price lunch (FRL) eligibility rates predicted their SAT scores.

How did this work?? 1) GOSA first identified public schools that had 20 or more SAT test takers for each of the past three years.
2) Every school's SAT score and FRL rate for each of the past 3 years was included individually, and their most recent SAT scores (critical reading + math) were used (n=951).
3) Schools' actual 3-year SAT averages were then compared to the scores predicted by their 3-year FRL rate average and the regression equation.

1200.00

$$

1100.00
1000.00
SAT
900.00
800.00

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

$

700.00

0.00

25.00

50.00

75.00

FRL

The results demonstrated that SAT scores were highly correlated (-0.791) with FRL eligibility; the higher a school's percentage of students eligible for free or reduced lunch, the lower its actual SAT scores were. This is not to say that poverty "caused" lower test scores, but only that a clear relationship exists between SAT scores and FRL rates.

District Richmond DeKalb Rome City Atlanta City Gwinnett Bleckley Richmond Whitfield Muscogee Fulton DeKalb Carroll Marietta City

School Davidson Magnet School Chamblee High School Rome High School Grady High School Norcross High School Bleckley County High School Westside High School Southeast Whitfield County High School Columbus High School North Springs High School Lakeside High School Bowdon High School Marietta High School

* This paper defines "significant" as 1 standard deviation (82 points) or more.

Who are Georgia's homerun hitters?
This model only estimates how well a school would be expected to perform based on three years' worth of SAT and FRL data. But just because SAT scores are associated with FRL eligibility does not mean that schools with many poor students cannot or did not perform well. In fact, 13 high schools in Georgia significantly outperformed their predicted scores.*
As one might expect, Davidson, Chamblee, Grady, Norcross, Columbus, and North Springs High Schools offer magnet or other programs that may attract higher- performing low-income students. This likely helped them outscore the prediction.

What helped boost those SAT scores?
It may be more useful, then, for education stakeholders to look critically at the techniques used by the non-magnet schools to increase their students' SAT scores. Such information will enable stakeholders to make inferences about effective preparation.

The SAT program specialist at the Georgia Department of Education, Becky Chambers, will research the preparation techniques used by the over-performing high schools. Her findings will be published in the next GOSA electronic bulletin so that those strategies can be replicated by all high schools.
Georgia's SAT Needle
In summary, a number of high schools in Georgia are exceeding national SAT averages and are not accepting poverty as an excuse for low achievement. As the rest of the nation's SAT scores have dropped over the past five years, Georgia's students have consistently increased their math and critical reading scores. While there is still room for improvement in all schools, the Governor's Office of Student Achievement commends those that are high performers, exceeding expectations, or showing large gains in student achievement.

For more information, contact the Governor's Office of Student Achievement, GOSA@gov.state.ga.us.