“The data that counts, and the data that really counts.” (UPDATED)
**This post is a work in progress; keep checking it for more information!**
During his walk to D.C., Dr. Jesse Turner shared some eye-opening information about what the data say about recent school “reform” in America. (Note: These are their data, measured largely by standardized tests– we’re not even talking about what is revealed when we take a broader look at what it means to be an educated/happy/successful person in a free society.) Despite the billions of dollars, time, and energy invested in the school reform strategies and test-based accountability practices mandated by the federal government, there has been little to no discernible benefit to schools or students. (We have, of course, lots of documented harms…read the rest of the blog!)
Re: Flat growth on the NAEP:
- Reading for 9-year-olds, 13-year-olds, and 17-year-olds.
- These graphs show that, from 1971 to 2008, average reading scores increased by 12 points among fourth graders, 5 points among eighth graders, and 1 point among twelfth graders.
- Since the NCLB reforms, the rate at which fourth grade reading scores rose, as well as the rate at which the racial score gap narrowed, slowed except for Latino/a 17-year-olds*. Note, too, that the test format was revised in 2004.
- From 1999-2004, nine-year-old reading scores increased by 7 points, whereas after 2004, they rose by 4 points. For 13-year-olds, there was no statistically significant change from ’99 to ’04, but a 3-point gain on the new test, and for 17-year-olds there was a 3-point decline just prior to NCLB, and a 3 point gain after.
- Between 1999-2004, the Black-White gap (hence, “BWG”) for nine-year-olds shrank by nine points, but only by 3 points after NCLB. The Latino/a-White (hence, “LWG”) gap for nine-year-olds shrank by 7 points just before NCLB, and 4 points after. For 13-year-olds, the BWG shrank 7 points before NCLB versus 4 points after, while the LWG widened 1 point before NCLB and 2 points after. For 17-year-olds, the BWG closed by 2 points before NCLB and widened by 2 points after, while the LWG widened by 5 points prior to NCLB and widened by 4 points after.
- Overall achievement levels and gaps for Black 17-year-olds have not recovered from their 1988 peak. The same is true of Latino/a 17-year-olds and their 1999 peak. Several notable trends during and immediately after this time period include:
- anti-plaintiff decisions in several desegregation cases concerning Black children beginning in the early 90s (for more on the legal impact of lower court actions and Supreme Court cases Dowell (1991), Pitts (1992), & Jenkins (1995), see Dr. Gary Orfield’s chapter in the 1999 volume Law & School Reform, edited by Dr. Jay Heubert)
- anti-plaintiff action in a significant desegregation case concerning Latino/a children (Keyes) in 1995
- the 1998 adoption of Proposition 227 (which banned bilingual education) in California, with a similar initiative succeeding in Arizona (2000) and the expiration of the Bilingual Education Act in 2002.
- Math for 9-year-olds, 13-year-olds, and 17-year-olds. (Not included in Dr. Turner’s presentation, but interesting nevertheless.)
- These graphs show that, over the same period of time (1971-2008), math achievement (as defined and assessed by the NAEP) has increased 24 points among fourth graders, 15 points among eighth graders, and two points among twelfth graders.
- As with reading, the rate of increase for overall scores and progress in closing the score gap between racial groups have slowed since 2004:
- For nine-year-olds, the average rate of increase was around 2 points per year between 1999-2004 (9-point gain overall); between 2004-2008 the rate was just 1 point per year (4-point gain overall). For 13-year-olds, students gained 5 points between 1999-2004, compared to 2 points between 2004-2008 (1 point per year vs. 0.5 points per year). For 17-year-olds, there have been no statistically significant changes since 1986.
- Regarding score gaps, the trend is flat to negative though the changes in the gaps aren’t statistically significant. The gap between Black and White nine-year-olds grew by 2 points in 2004-2008, after shrinking by 5 points in the period before, while the gap between Latino and White nine-year-olds didn’t move at all between 2004-2008, after shrinking by 8 points in the period before. For 13-year olds, the BWG shrank by 2 points between 2004-2008 after shrinking by 5 points between 1999-2004, while the LWG didn’t change at all in the most recent time period, compared to closing by 1 point between 1999-2004. For 17-year-olds, the BWG closed by a point in 2004-2008 compared to closing by 3 points between 1999-2004, while the LWG widened by 2 points in both time periods (no change).
Re: The effects of Reading First
The National Reading Panel’s 2000 report (PDF) examined several areas of reading instruction and looked at key questions related to how to help all children learn to read well, and to help struggling readers overcome reading difficulty. The document* stated that one size does not fit all when it comes to reading instruction, and encouraged a balanced approach to literacy instruction.
The federal Reading First program that followed was supposed to help schools intervene in the early years (K-3) to prevent future reading problems. The government mandated that schools getting funds from the program use certain curricula and assessments (ones the Bush Administration had approved as being “scientifically-based”). There has been a lot of controversy over these programs. Notably, the impact study commissioned by the Department of Education itself found that Reading First had no positive impact on reading comprehension.
- Interim Report (April 2008): In the sample schools examined from 2004 to 2007, Reading First increased the amount of instructional time spent on the five essential components of reading* promoted by the program, but students in Reading First schools showed no improvement in comprehension.
- Final Report (November 2008): Same finding– no improvement in comprehension, though time spent on elements of the program was shown to increase.
*Note that there is also a controversy over the report itself, and its identification of the so-called “five essential components of reading.” The NRP Minority View (PDF) submitted by panelist Dr. Joanne Yatvin stated several objections to the implicit assumptions and claims made in the official report. The document is an incredibly important read for anyone interested in understanding the unavoidable limitations of the Reading First program. The official report has been misused to validate a very narrow view of what reading is and what “scientifically-based” reading instruction looks like, to the detriment of millions of students.
This pamphlet (PDF; needs to be rotated or printed for easy reading) prepared by Dr. Yatvin, Dr. Constance Weaver, and Dr. Elaine Garan also offers succinct, point-by-point explanations of why many people are concerned about Reading First, including the fact that the National Reading Panel never recommended the commercial and scripted reading programs that became a part of Reading First, and that these programs haven’t been shown to be effective. There is more information on what the reading research does say here.
Re: College remediation rates:
One key justification for many of the No Child Left Behind and Race to the Top reforms is the idea that students are arriving at college unprepared to do college-level work. But has NCLB/RTTT-style reform reversed those trends?
- Not in Colorado:
- Only three of Colorado’s ten largest school districts saw a decline in college remediation rates for their graduates, and these are relatively more successful districts that had comparatively lower rates to begin with (Jefferson County, Boulder County, and Fort Collins)– not the kind NCLB was most intended to help. Statewide, there is virtually no change; since 2005, the overall remediation rate increased 0.04% (around 32% of Colorado students qualify for remedial courses in college, and they disproportionately come from the kinds of schools and districts targeted by NCLB).
- Since 2005, Adams 12* saw their rate increase by a little over 2% (for an overall figure of 38% of students qualifying for remediation) Denver’s rates are up almost 6% (55% overall), and Aurora Public Schools rates are up over 11% (56% overall). See the rest of the data here (same as previous link).
- *The Adams 12 5-Star School District is separate from the smaller Adams 50 and Adams 14 school districts, though they are all in the same county. While around 40% of Adams 12′s students are nonwhite and 31% are eligible for free and reduced price lunch (FRL), 78% of Adams 50 students are nonwhite, and 71% are FRL eligible, and 82% Adams 14 students are nonwhite, with the same percentage being FRL eligible . I’m currently looking for the remediation rates for Adams 14′s and Adams 50′s high school graduates.