The Marin Independent Journal celebrates the test scores of local children, but misstates the relationship between test scores and SES:
Marin students performed better in standardized tests than students in other Bay Area counties and were more likely to take Scholastic Aptitude Tests than their peers, according to a report to be released today...
In 2003, Marin had the highest English-language arts score in California Standards Tests in the Bay Area, with 45 percent of students showing proficiency. Algebra scores also beat those from surrounding counties at 45 percent proficiency...
Marin's success is due primarily to the commitment the county has made to educating all students, according to Mary Jane Burke, superintendent of the county Office of Education.
In the past, national studies have shown that high test scores are, more than anything else, the product of socio-economic status of families, with students from the high end of a continuum consistently scoring higher on standardized academic achievement test than those from the low end. This is borne out in Marin, where a high percentage of parents of students tested are educationally and economically on the high end of the scale, compared to the rest of the state, as well as the nation.
Correlation does not imply causation. Simply because, overall, students who come from wealthier families may do better on tests, that doesn't mean that, as stated above test scores are caused by the wealth of the parents.
It is logical to draw that conclusion in our society, because kids from wealthier families have more opportunities and tend to go to better schools. But their parents might be wealthy because IQ and income tend to be correlated in our society as well, and it might be that intelligence - and learned traits like self-discipline and competitiveness - account for both parental income and child test scores. How many parents in Marin County had to work long hours at hard jobs to earn the money to live there? Who's to say their kids didn't inherit or learn those traits from them?
This might seem like a nitpicky point, but it's not. One of the most important concepts taught in Stats 101, is that correlation between A and B does not imply that A causes B or B causes A, because a third factor C might cause both to occur. This concept is essential for research because missing that third factor can lead to totally erroneous conclusions like "Poor kids have no chance of making good test scores" or "Tests don't measure anything other than parental income," both of which assume that money in and of itself is the most important factor affecting test scores.
At least the article avoids the truly boneheaded statements, and it doesn't go into the nonsense about how it's not "fair" for some kids to have more money than others, but the author still gets it wrong by assuming that correlation implies causation. Congrats to Marin's students, though (who I bet would be rather offended if you told them they only scored well on tests because of Mommy and Daddy's money).
Update: As Devoted Reader Jeff W. points out, I missed the real story:
Marin is the wealthiest county in California. Kids there enjoy advantages that most kids in this country can only dream of.
Yet only 45% were proficient in English and algebra. That's the real story.
According to Wikipedia:
The median income for a household in the county is $71,306, and the median income for a family is $88,934. Males have a median income of $61,282 versus $45,448 for females. The per capita income for the county is $44,962. 6.60% of the population and 3.70% of families are below the poverty line.
Can't blame poverty for poor school performance, can they? Here is the county's STAR score report for 2002. According to this, the scores for Language Arts breaks down as follows:
Grade ----- % scoring Proficient or Advanced
2 ---------- 56
3 ---------- 61
4 ---------- 67
5 ---------- 61
6 ---------- 60
7 ---------- 66
8 ---------- 63
9 ---------- 58
10 --------- 55
11 --------- 52
For each grade, the percentage scoring only Proficient (not advanced) is anywhere from 24% to 43%. So I'm a little confused. Is the 45% for 2003 reported in the article above for only the Proficient category (which would reflect a small increase) or for both Proficient and Advanced together (which would reflect a large decrease)?
Either way, Jeff is right. I missed the real story, which is (a) that in 2002, with the data I'm reporting here, only 52% of 11th-graders in a very wealthy county scored at or above Proficient in Reading, and (b) sad as that is, they're still doing better than the country at large, because 57% of Marin's 11th-graders scored above the national 50th percentile mark.
Posted by kswygert at February 26, 2004 11:18 AM