Math scores on the statewide grades 3-8 exams suggest that math instruction in the city schools has turned a corner and gained real traction.
More than 75 percent of third graders met standards, almost 71 percent of 4th graders and 61.3 percent of 5th graders also did so, and the middle school grades demonstrated real progress, though they have a ways to go.
Commissioner Mills, in his press conference, singled out New York City results. He called them “encouraging” and pointed out that the city is performing at or above the level of the urban/suburban districts in the state and that its Grade 3-8 average of 57 percent meeting standards is “closing in on the statewide” average of 65.8 percent.
Chancellor Klein, in his press conference later in the day, produced charts showing that in the last four years 4th grade performance has increased by 19 points in the percentage meeting standards and 8th grade is up 9 points over four years. Those gains are double and triple the gains in the rest of the state.
There was an increase in Level 1 students in 4th and 8th grades, but the tests were harder this year, according to Mills, and comparisons with last year are not “apples to apples.” For example, some previous 5th grade content appeared on the 2006 4th grade exams, and similarly with other grades.
Previous-grade teachers had not known of the new content until too late and so many teachers last year had to teach 10 months of math in 7 months, UFT Teacher Center math specialists said. They performed admirably.
Randi Weingarten’s press statement welcomed the results but pointed out some of the sticky areas. “Overall, the city’s math scores are moving in the right direction. The increase in the Level 1s – those children who score lower than basic minimum standards – and the drop in middle school scores, particularly in 7th and 8th grades, are troubling. Lower class sizes, redoubling our focus on safety and addressing adolescent social development needs would hugely help – and could be funded with CFE money,” she said.
The pattern of lower middle-school performance is not limited to New York City. In fact, it is worse by far in other big cities in the state, and occurs throughout the state. Efforts to address this have demanded exceptional effort from math teachers.
In the city, 8th grade students who took the assessment in 2006 were the first class to have used IMPACT Math for 3 years. They and their teachers were the first to implement the Impact course 1 in 6th grade, the first to implement course 2 in 7th grade and the first to implement course 3. This is a very rigorous program, according to Teacher Center staff, at the end of which students have completed a year of Algebra. In addition, the new standards were not all addressed in IMPACT. There were gaps that teachers had to find by themselves, and find the resources to teach them.
The instruction requires highly skilled teachers, especially as the standards are ratcheted up. Mills called for increased professional development to boost math performance.
Weingarten noted that in Region 5, where the Teacher Centers supply all the math professional development, the gains were the best in the city, as they were for ELA.
In his press conference, Chancellor Klein said he is focusing the lead teacher program on the city’s middle schools. But he said the math results overall show the “real sustainability of the work we are doing.”
If you are a math teacher reading this post, treat yourself to something. You deserve it, and we thank you.
3 Comments:
1 R. Skibins
· Oct 15, 2006 at 12:32 am
Math scores would increase even further if we could ditch the woeful “Everyday Math,” which has been condemned by experts throughout the nation.
2 jd2718
· Oct 21, 2006 at 10:20 am
Hypothesis 1: The schools are doing a better job
Hypothesis 2: The difficulty of the tests varies widely year to year.
We can begin to test these hypotheses by looking at cohorts instead of at individual grades. What happened to the 3rd graders of 2003? They became 4th graders in 2004, 5th graders in 2005, and 6th graders in 2006. Their scores (listed by their grade level in 03):
GL03 – 03 – 04 – 05 – 06
K.g – na – na – na – 75
1st – na – na – 65 – 71
2nd – na – 58 – 78 – 61
3rd – 52 – 68 – 54 – 53
4th – 67 – 39 – 41 – 44
5th – 39 – 40 – 40 – 39
6th – 31 – 34 – 41 – na
7th – 28 – 42 – na – na
8th – 34 – na – na – na
There is not enough data here. There must be better break-outs. But look at each group of kids. Of the 5 cohorts with at least 3 years of data (2nd through 6th grade in 2003), do you see any trend to improvement for a group of kids?
These numbers bounce around, trendless. We are looking at artifacts of unreliable tests, not of changes in student acheivement.
Jonathan
3 Maisie
· Oct 25, 2006 at 10:14 am
Yes, we are looking at unreliable tests, or not exactly unreliable but they are not equated. Some were city tests, some were state, developed for different purposes by different publishers. And this year’s 3-8 state tests are all new. I agree with Jonathan that looking at cohorts of kids longitudinally is a far better measure than “snapshot” looks, but we don’t yet have reliable longitudinal data. That will take another couple of years (assuming that the state does an adequate job of equating the tests year over year).