Fewer South Carolina students took the SAT in 2010 than in 2009, as participation rates for African-American and white students fell. This finding raises questions about the S.C. Department of Education’s claim that “South Carolina is doing dramatically better with access … as more minority and low-income students are aiming for college.”
Following the news that South Carolina’s SAT scores saw a five-point drop in 2010, the S.C. Department of Education (DOE) issued a press release asserting:
-
Participation, or “access” to the SAT has increased (which would seem to explain the drop in SAT scores)
-
SAT scores for public school students are not down by five points, but by two
-
Advanced Placement (AP) testing scores are up
The implication is that the drop in SAT scores—the fact that South Carolina’s scores are the second-worst among all 50 states—is “no big deal.” This is because participation is up, which would tend to push scores downward as students who don’t usually take the SAT are now doing so. And because public school students seemed to do better than everyone else, as their scores were down by two points instead of five—not to mention AP scores having increased.
Participation is Down
AP scores aside, the DOE’s claims about the SAT are misleading. Most important, overall participation is not up, it’s down. The total number of statewide test takers decreased from 25,217 in 2009 to 25,122 in 2010.
The only way the Department of Education can claim participation has increased is by selectively counting certain categories of self-identified students. This is how it works:
-
The SAT gives students the option to identify the school they attend and then uses this information to classify students as either public, private or “Other/Unknown.”
-
There were 25,122 test takers in South Carolina. Of these, 81.7 percent identified themselves as public school students, 9.8 percent as private school students. That leaves approximately 8.5 percent of students who either do not attend a public or private school or simply failed to self-identify. These students are classified by the College Board as “Other or Unknown.” Homeschoolers are included in the “Other” category.[1] Presumably, the “Unknown” category includes a proportional representation of all types of students, which would indicate that most of these students are public school students.
-
The number of test takers who identified themselves as public school students increased by about 5 percent from 19,534 in 2009 to 20,530 in 2010. Likewise, the number of 12th graders in the S.C. public school system increased by about 1,172 students between 2009 and 2010.
-
The number of test takers self-identified as attending “independent” or “religious affiliated” schools declined by 66 students, or 2.6 percent, from 2009 to 2010. Likewise, enrollment in private schools has declined, because of the current recession.[2]
-
Students from the “Other/Unknown” category decreased from 3,158 in 2009 to 2,133 in 2010: that’s 1,025 students—a 32 percent drop.
In other words, the overall number of statewide test takers declined as did the number of self-identified private school test takers, as well as students classified as “Other/Unknown.” But, at the same time, the number of self-identified public school test takers increased.
Likely conclusion: The increase in self-identified public school test takers is due to more public school students identifying themselves as such—not due to more students actually taking the test.
Overall, it’s thus safe to say that fewer public school students took the SAT this year than last.
African American and White “Access” is Down
As for minorities and low-income students, the data shows mixed results when we look at all students statewide.[3]
SAT Participation Rates by Ethnicity and Income: 2009 to 2010
|
Demographic Type
|
2009
|
2010
|
Change?
|
|
White
|
16302
|
16172
|
Down -130
|
|
African American
|
6571
|
6364
|
Down -207
|
|
Hispanic
|
624
|
758
|
Up 134
|
|
$0-$20,000
|
2199
|
2231
|
Up 32
|
|
$20,000 – $40,000
|
3282
|
3267
|
Down -15
|
|
$40,000 – $60,000
|
3099
|
2994
|
Down -105
|
|
$60,000 – $80,000
|
3106
|
2776
|
Down -330
|
As the table above indicates, 134 more Hispanic students took the SAT in 2010. But 207 fewer African-Americans took the test in 2010. So, overall “access” declined for the state’s two largest minority/ethnic groups. Granted, 32 more students from households earning $20,000 or less took the SAT in 2010. But also note that 15 fewer students from households earning $40,000 or less took the test in 2010.
Test Scores Down by 5 or by 2?
Finally—and this bolsters our conclusion that most of the supposed increase in participation is due to a shift in identifying “Other/Unknown” students as public school students—we should say a word about the DOE’s claim that public school SAT scores are down by two points, rather than five.
According to the department, “South Carolina public school seniors’ average composite score for critical reading, math and writing was 1,443, down two points.” Yet, when we include all test takers, the statewide average is down by five points from last year. At the same time, scores for students from private schools and religiously affiliated schools increased—by 20 points and 1 point, respectively.
The numbers don’t add up—that is, until we account for those students in the Other/Unknown category. Unfortunately, no one wants to claim these students because their SAT scores are the lowest among all three categories of students.
-
Students in the “Other/Unknown” category scored an average of 1,372, compared to a statewide average of 1,447 and a national average of 1,509.
-
While a fraction of students in the “Other/Unknown” category are homeschoolers, preliminary data indicates that S.C. homeschoolers scored an average of 1,572 on the 2010 SAT—63 points above the national average. [4]
-
Recall, also, that students from religiously affiliated schools scored an average of 1563 while students from independent schools scored 1528.
Thus, test takers from the Other/Unknown category account for the three-point difference between public school test takers and all test takers statewide. Yet, as we have already seen, most of these Other/Unknown students are actually public school students.
Conclusion
More troubling than any selective reading of state SAT scores, is the impression that the Department of Education is content with lower test scores because “access” has improved. The promise of the S.C. public education system is that all students are being equipped with the skills they need to excel in life. More SAT test takers does not translate into success when the state’s average score is the second-worst in the country.
Clearly, the Department of Education needs to shift its focus to student performance. Doing so requires taking the emphasis off of equating more spending with better results and reconsidering policy reforms that could revolutionize South Carolina’s public school system. As we’ve written elsewhere, some of these reforms include weighted student funding and school choice. But the DOE has consistently opposed these ideas.
What, then, is their plan? Simply improving access is not a solution—especially when all they mean is access to a failed system.
Nothing in the foregoing should be construed as an attempt to aid or hinder passage of any legislation.
[1]As of publication, the exact percentage of homeschooled test takers was not available, but homeschoolers are able to specifically self-identity themselves as such. For general reporting purposes, however, they are included in the “Other/Unknown” category.
[2]Exact data for the 2009-2010 school year not available as of publication. That said, the nationwide trend for homeschooling students is increasing. But homeschoolers, generally, have higher than average SAT scores.
[3]Data specific to public school test takers not yet available as of publication.
[4]As reported by the S.C. Association of Independent Home Schools, the state’s only home school association formally recognized in S.C. law. These numbers are derived from 36 test takers and do not include all home school test takers statewide. If these numbers are any indication, however, the number of homeschoolers included in the Other/Unknown category is clearly very low.