Gary Rubinstein’s Blog: Pay No Attention to the Falling Tennessee Reading Test Scores
To ‘reformers,’ particularly ones in Tennessee, only two things matter in education: Grades 3-8 Math scores and grades 3-8 Reading scores. This was what got them the shout out from The President of The United States in the last State of the Union address. Tennessee had the greatest combined 4th and 8th grade math and reading gains of any state.
When it came out later that their 12th grade NAEP scores did not have these gains, they said that the 12th graders didn’t have the opportunity to get the full ‘reform’ treatment that the younger students got.
The Tennessee state tests are called the TCAPs. This year there was a fiasco when the education department, led by former TFA VP and former husband of Michelle Rhee, Kevin Huffman, announced that they were going to be late with the test scores. This raised a lot of suspicion that maybe they were thinking of ways to spin this data. I’m not sure if this is what happened since in Tennessee, they can spin data very fast so I don’t think this would cause the delay. Still, there was a delay which at least made them demonstrate that they could not meet an important deadline.
So the scores came out this week and, in true Louisiana style, they are celebrating the results. But in the celebration they are downplaying the two most important tests of all, 3-8 math and 3-8 reading. For a few years those scores were steadily increasing. Reformers always assume that score increases can continue year after, when, of course, they are bound to level off or even go down. This is what happened this year on the two big results.
3-8 Reading dropped from 50.3% to 49.5% while 3-8 Math increased from 50.7% to 51.3%. Now these are small changes and not tremendously ‘statistically significant’ either way. But they are pretty ‘flat’ which is an embarrassment to them. Here is the graph showing the past 5 years of scores for Math and Reading. You math geeks out there will understand when I say that the second derivative of their test scores graph is negative. (Note: They also had a Science bar graph showing steady improvement. It was there, I figure, to draw attention away from the lack of progress on the big two scores, so I removed it.)
Looking at the scores broken down by grade, one irony is that the largest drop of any grade in any subject was 3rd grade reading which dropped from 48.8% to 43.8%, a drop of 5%. This might be statistically significant, but more importantly, this is the group of students who had the most opportunity to benefit from the reforms put in place in Tennessee, so ‘reformers’ should expect that group of 3rd graders to outperform previous groups. Third grade math also dropped from 59% to 56.5%.
I don’t want to make too much about test scores since they are controlled by the state, with cut scores and things like that and they can easily manufacture a miracle next year and say that 2014 was an anomaly. I suppose the national common core tests, which TN recently pulled out of, would have one benefit in that it wouldn’t be as easy for states to manipulate the results. There are other, much less costly, ways to ensure that state tests are of comparable difficulty. I’m pretty good at looking at a math test and judging how difficult it is. Maybe state tests can go to some impartial people who can expertly gauge the difficulty of the test. I haven’t worked out all the details yet. Certainly there are tests like the SAT, AP, and ACT which are national tests and states can lie all they want on their own tests, but eventually they will have to explain why their SAT, AP, and ACT scores are so low, like in Louisiana.
‘Reformers’ are slick. They get a whole bunch of data and focus on the things they want. The next year they can focus on completely different things if it makes them look good.
This blog post has been shared by permission from the author.
Readers wishing to comment on the content are encouraged to do so via the link to the original post.
Find the original post here:
The views expressed by the blogger are not necessarily those of NEPC.