Posted by
Jessica Hughes on Tuesday, February 03, 2009 3:03:21 PM
While I was happy to read in the Washington Post of the admitted success of School Choice in Washington D.C., I couldn't help but notice the way the Washington Post drew a causation determination where none could legitimately be drawn.
This particular practice is brilliantly addressed by Thomas Sowell in his book "The Vision of the Annointed (Self-Congratulation as a Basis for Public Policy)." As anyone who watches the news has seen, when there is a correlation between circumstances A and B, we are frequently told that either A caused B or B caused A. However, as Sowell wisely points out, there are further possibilities:
- A causes B
- B causes A
- Both A and B are caused by C
- The correllation is coincidental
Buried in the middle of the article (
Full Text) is the following passage:
- Friendship Public Charter Schools -- the city's largest charter network, with five schools and more than 4,000 students -- has a surplus of $3.4 million that has funded cutting-edge equipment, including computerized interactive whiteboards that are found even in preschool classrooms.
The extra funding, it turns out, coincides with improved academic performance: The schools with the largest surpluses have ranked at the top on test scores. -
Note the switch: "Extra funding" magically becomes "Largest Surpluses". These are arguably two distinct issues. The author of the piece obviously intends to convey that more funding equals better academic performance and from this limited sampling that conclusion could be drawn.
Out in the real world, though, there is an alternate sampling that might belie the conclusion drawn here, that sampling including the public schools in the same neighborhoods with the same student population, with the highest per capita spending of any public school system in the nation and one of the worst academic results.
Enter the "Largest Surpluses" switch. Although the conclusion drawn was that the overall amount of funding correlated to better academic performance, the evidence given for this conclusion was that the schools with the largest surpluses had the best performance. Arguably, the schools with the largest surpluses may have had less overall funding but spent their available funds more responsibly. This air of responsibility with funds could indicate a propensity on the part of the administration of such a school to demand high results relative to spending.
While I cannot prove such a thesis with the available information, neither can the Washington Post prove that more funding equates to better academic performance based on a correllation between performance and surplus. However, my own thesis at least does not stand in defiance of the available evidence as does theirs when the public schools of Washington D.C. are considered as part of the sample.