Commenters on the AJC Get Schooled blog were largely skeptical of former Stanford education Dean Marshall S. Smith’s essay today noting Atlanta’s impressive gains on a respected federal test known as the Nation’s Report Card. While Smith says cheating on the NAEP tests is highly unlikely and the APS gains are credible, blog readers were dubious. Here is a sampling of their comments:
Former APS Student: As for NAEP, it is safe to say that the scores show that in the mist of all the chaos, cheating and scandals, there was actually learning happening in Atlanta schools. Even though thousands of kids may have been affected by systematic cheating, tens of thousands benefited from some sense of accountability. I believe the majority of teachers taught the students. The students attending the Atlanta Public Schools now are probably the highest-performing students that have ever attended school in the system's long history.
Centrist: Marshall S. Smith could have written a single sentence to sum up his conclusion: "Pay no attention to that man behind the curtain."
Astropig: Statistics are like political prisoners — torture them enough, and they will confess to anything
Longtime Educator: There is no doubt that large-scale cheating went on at APS and was tolerated and promoted by the leadership of the district. That is fact. When given the unusual scores on the NAEP, a smart person would not scratch his head and say, "Gee, maybe this means they did not cheat after all." A smart person would say, "These really high scores make Atlanta look better than most urban areas, but we know they are willing to lie and cheat to look good, so we need to really examine these scores, because we know the system is dishonest." I think the answer will be found in the way the district selected the representative sample of students.
Title One: Because NAEP doesn't test every student or even every school in a state and its protocol is less than transparent, it can hardly carry the weight of a state-administered test like the Georgia CRCT or a norm-referenced test like the Iowa Tests of Basic Skills.
Jerry Eads: Having run another state's testing program for a few years and hence having been responsible for coordinating that state's NAEP participation, my take is that it's just about impossible to "cheat" on NAEP. The NAEP folk watch pretty closely. That's not to say it couldn't be done, but it would take some really determined people making a very concerted effort. I think it unlikely that a certain district's leadership was sufficiently capable to pull it off. In contrast, it's relatively simple to manipulate outcomes on state minimum competency testing. It's even simpler to do it stupidly and get caught. As we know, thanks to the AJC's work, it has been quite common across the country for a long time. As long as we maintain our naive and shortsighted punitive test-based approach to school accountability, it will continue. Guaranteed. The only way we'll stop it is to develop sane policy that treats teachers as professionals creating citizens, and not indentured servants making widgets.
