High School Reading Remediation — A Tale of Teachers’ Success and Management’s Failure
Wheat Ridge High School (WRHS) is, by many measures, the most diverse high school in Jefferson County, Colorado, the nation’s 36th largest school district which each year spends a billion taxpayer dollars on its public schools.
WRHS offers its 1,200 students SPED, gifted, STEM, STEAM, and English Language Learner programs, in addition to general education. Fifty percent of its students are eligible for free and reduced lunch (the actual number of economically distressed families is higher), and nearly fifty percent are minorities.
Over the past year, the Wheat Ridge community has experienced growing conflict with the district’s head office, which I have described elsewhere (see, “Diversity Be Damned: Progressive Hypocrisy in Jefferson County, Colorado”).
In the next school year Jeffco plans to significantly reduce funding to WRHS, when it withdraws $150,000 that the district has, since 2009, provided to the school to pay for its only GT Center High School program. This was a highly controversial decision, as the provision of adequate education to both special education and gifted children is mandated under Colorado’s Exceptional Children’s Educational Act.
At the direction of Jeffco’s new superintendent Jason Glass, the district recently formed a specially selected team of administrators and a few GT parents to resolve the school’s budget issues. In the context of Colorado state law, this was, at minimum, highly unusual, in that the legislature had, almost a decade ago, created independent bodies — School Accountability Committees — to review achievement results and provide advice to principals and districts on budget issues.
This rump group (which the district termed a “Rapid Improvement Event” team) recommended a significant reallocation of Wheat Ridge’s reduced funding in order to support that part of the school’s GT Center Program which serves GT students with the greatest need for additional supports (many of whom have been evaluated as both special education and gifted, or “twice exceptional”).
The RIE process culminated in the Superintendent’s decision to approve this reallocation without any involvement of or input from, much less concurrence by, Wheat Ridge’s School Accountability Committee. Indeed, when the SAC met to discuss the RIE’s results with district staff, Scott Allensworth, the Achievement Director responsible for WRHS, told the group that the funding decision was final, and all the SAC could do was ask questions of clarification about it.
It is thus quite clear that the RIE process violated Colorado’s School Accountability Law, and utterly disregarded the powers and duties of the SAC as established by the state legislature.
Yet the RIE process had two even more egregious shortcomings.
The first was the failure of district staff to disclose material information to the RIE team. Specifically, they failed to disclose Jeffco’s subsequently announced plans to open a new GT Center Program at Arvada High School, just five miles away, as well as a new Arts High School even closer to WRHS.
Both of these could reasonably be expected to have a material negative impact on the projected Wheat Ridge enrollment data that formed the basis of the RIE team’s quantitative budget deliberations.
The district’s argument to the RIE was essentially that the $150,000 in historical district funding for the WRHS GT Center program could be replaced with Student Based Budgeting funds by drawing down the school’s budget reserve to about three percent of revenue.
Since the launch of SBB, WRHS has run an annual surplus above the district’s recommended level of five percent of revenue. This was because the school’s enrollment was always on the cusp of the SBB “cliff” where a slight increase in the number of students would trigger the elimination of the “size factor”, which is additional SBB funding for so-called “small schools.” Under such circumstances, it was prudent for WRHS to use its surplus for one period expenses, but not for expenses (like staff hiring) that would continue over multiple periods (i.e., where loss of the small school funding could force the firing of staff who had just been hired the year before, and elimination of the classes they taught).
Had the district’s plans to open both an arts high school and a second GT center high school been disclosed to the RIE team, somebody might have asked how many students WRHS would have to lose before its surplus was wiped out. The answer is 41 — or just 3.25% of its SY 2017/18 enrollment. Whether such a loss would trigger district demands for staff cuts to restore reserves to five percent of revenue is unknown.
Would the RIE team have made the same recommendations if they had known about the district’s plans to open the competing programs so close to WRHS? We’ll never know.
In light of Dr. Glass’ and the Jeffco Board of Education’s repeated claims of their deep concerns with “equity” and closing achievement gaps, the most galling aspect of the RIE process was the utter failure of the RIE team to consider the adequacy of current funding levels for reading remediation activities at WRHS.
In today’s rapidly evolving, knowledge-based economy, lifetime learning — which is predicated on the ability to read well — is clearly critical. And the future price that will be paid by students who cannot read well by time they graduate (if they do) is truly frightening.
This is why keeping children at or above grade level reading standards throughout their K-12 journey is so important. When districts let students fall behind, research shows that reading remediation at the secondary level is a difficult process that is time and resources intensive.
For example, here is the conclusion from a recently published meta-analysis of existing research in this area: “The findings from this synthesis provide compelling evidence that accelerating reading growth in the upper grades may be more challenging than in the earliest grades, even when extensive interventions are implemented…For secondary students with significant reading difficulties, very intensive and sustained interventions may be required to maintain reading growth each year of school” (“Extensive Reading Interventions for Students with Reading Difficulties After Grade 3” by Vaughn et al).
Unfortunately, Wheat Ridge High School has a significant percentage of students who arrive for freshman year reading far below grade level. This year, it topped 70%.
Going back to the 2015 CMAS (PARCC) ELA assessment, 44 WRHS 9th graders were more than approximately two standard deviations (SDs) below the cut score for meets/exceeds state standards, 60 were between 1 and 2 SDs below the cut score, and 79 were less than 1 SD below the cut score (CMAS cut score intervals don’t exactly line up with standard deviation intervals, but they are close enough for this analysis).
It is important to put these standard deviation differences into context. To facilitate comparisons between studies, the amount of achievement improvement produced by various interventions is often specified in terms of the number of standard deviations. This metric is also called “effect size.”
The average annual effect size for secondary reading improvement interventions is .21 standard deviations. Note that this average is mostly based on the results of middle school interventions. The relatively few studies that have been done on high school reading remediation have generally found significantly lower results.
Note also that .21 SDs is the average effect size of interventions over and above the average grade-to-grade improvement in reading skills (for both 9 to 10, and 10 to 11, this is .19 standard deviations per year; for 11 to 12 it is .06 — see, “Performance Trajectories and Performance Gaps as Achievement Effect-Size Benchmarks for Educational Interventions” by Bloom et al).
For example, consider the 79 WRHS students who, at the end of ninth grade in 2015, were less than one standard deviation below the cut score for meets/exceeds state standards. Assume that, on average, they were .50 standard deviations below the cut.
Assuming WRHS had sufficient resources to provide them with reading improvement interventions of average secondary school quality in grades 10 and 11 (and note that this is an aggressive assumption for high school), and further assuming that these students also achieved the average grade to grade level improvements of .19 SDs per year in grades 10 and 11, then they might all meet the “college and career ready” standard for Evidence Based Reading and Writing (EBRW) on the SAT taken at the end of 11th grade. If effective interventions continued for another year, these students would have an even higher chance of meeting the C&C ready standard on the 12th grade SAT.
Now consider the 60 students who at the end of ninth grade in 2015 were between 1 and 2 standard deviations below the ELA cut score. Assume that on average, they were 1.5 standard deviations behind. Assuming the same .21 standard deviations impact of reading interventions, and again assuming that these students achieved the normal grade to grade improvement, then by the end of 11th grade, you would expect that on the SAT they would still score a full standard deviation below the EBRW cut score for college and career ready. By the end of 12th grade, they would still be about .80 SDs below (although they would certainly be reading significantly better than they were at the end of ninth grade).
In practice, between SY 14/15 and SY 16/17 students whose reading level was less than one standard deviation below the cut score for proficiency received reading interventions only in their regular social studies and English classes at WRHS. Students more than two SDs below the cut score received interventions via Special Education programs. Reading remediation specialist resources were focused on those students between one and two SDs below the cut score.
The following table shows the number of these students who attended WRHS between School year 2014/2015 and SY 2016/2017. For the earlier years, the numbers in the table corresponds to WRHS 9th graders who were only partially proficient in reading on the TCAP; for later years, the number who partially met the 9th grade CMAS ELA standards. Both of these metrics roughly equate to students who were between 1 and 2 SDs below the respective proficiency cut scores on TCAP and CMAS.
The average over this period was 349 students a year.
When ninth graders took their CMAS ELA assessment results in the spring of 2015, WRHS employed two reading remediation specialists, and had a coordinated focus on reading remediation in both its social studies and English classes. Moreover, the reading remediation specialists worked with the same students for multiple years.
Assume that each reading remediation specialist spent 6 hours per day with students, over the 175 contact days in the school year (i.e., days on which teachers are in contact with students). That yields about 2,000 hours per school year that were available for reading remediation outside the classroom.
This equates to 5.73 hours per student. However, to improve efficiency, students were grouped into classes of ten. This increased the number of hours available to each group to 57.1, or about 1.6 hours per week (about 2 class periods per week) over a 35-week school year.
Note also that in the 2016–2017 school year, WRHS launched CareerExplore, a pilot focused on 29 students who had lost hope, and were at danger of substance abuse, dropping out, and far more serious lifetime consequences. All of them were in the group served by the reading remediation specialists.
These students received a variety of supports, with a goal of helping them to each achieve three certified industry competencies, and successfully complete paid internships with construction, hospitality, or healthcare employers. Twenty-eight met this goal. One only achieved two of the three targeted certified competencies. And every student demonstrated a substantial increase in reading achievement growth.
Now let’s look at the reading remediation results that WRHS obtained between SY 14/15 and SY 16/17 from its focus on reading improvement in social studies and English classes, its investment in two reading remediation specialists, and, in the spring of 2017 the launch of the CareerExplore program. Note that the following estimates are based on publicly available data.
On the 2015 9th grade ELA assessment, WRHS average score was .42 SDs below the cut for meets/exceeds state standards. The participation rate on the test was 76%. On the 2017 11th grade SAT EBRW test, participation was 92%, and WRHS’ average score was .54 SDs above the cut score for college and career readiness. While not a perfectly matched cohort, this .96 SD improvement in results implies (if it is representative of the system’s normal dynamics) an excess two-year gain of .58 SDs over the normal grade 9 to grade 11 improvement of .38 SDs.
A critical uncertainty is how much of this excess gain represented the effect of remedial reading programs (both in and outside of class) and how much was driven by the 16% difference in testing opt-outs (i.e., strong students opting out of the 9th grade CMAS but taking the 11th grade SAT).
One way to think of this is in terms of breakevens. For example, if 1/3 of the gain was from the change in optouts, and 2/3 was due to effective reading remediation, WRHS would have an annual effect size for reading remediation of about .19 SDs, which is significantly better than the results for the few high school specific studies that have been published. In contrast, if 2/3 of the reading gain was due to the difference in optouts, the estimated effectiveness of WRHS reading remediation would decline to about .10 SDs per year.
So which is more likely? WRHS reading remediation specialists note that, when WRHS had two specialists, annual improvements of one additional grade level (above normal grade-to-grade growth) were regularly achieved for those students who were less than one SD below the cut score, along with gains of one half to three quarters of a grade level for students more than one SD below the cut score (the gains for students in the CareerExplore pilot were a grade level or more).
As previously noted, between 9th and 11th grade, the normal reading gain is .19 SDs per year. Combining this with the grade level improvements reported by the reading remediation specialists suggests that ascribing half the observed .29 SDs annual reading improvement — that is, about .15 SDs per year — is a very reasonable (and quite possibly conservative) estimate of the aggregate effectiveness of the WRHS reading remediation program when it was resourced with two specialists.
In the 2017–2018 school year, Wheat Ridge is down to just one reading remediation specialist to serve 276 students (based on fall MAP scores for 9th graders, and previous 9th grade CMAS results for grades 10, 11, and 12). Keeping remediation group size constant at 10 students, this has reduced annual specialist remediation time per group from 57 hours to about 36 — a cut of 37%.
This is sure to have a negative impact on reading remediation results at WRHS.
Contrast this with the alternative case in which WRHS still had two reading remediation specialists. If group size were held constant at 10 students, annual instructional time per group would be 71 hours, or more than two hours/week over a 35-week school year (an increase of 25% over the instructional dosage during the SY 14/15 to SY 16/17 period). Alternatively, group size could be reduced to 8 students, to increase instructional intensity while keeping the time dosage unchanged at 57 hours per year.
In combination with the CareerExplore program (which also suffers from a lack of district funding), both of these approaches would have been likely to deliver even greater student reading gains than those that WRHS has achieved in the past. But that is not to be.
The RIE team never asked about any of this. Either they didn’t care about those students who need reading remediation, or they just assumed that one reading remediation specialist is sufficient, despite the continuing failure of so many students to meet the college and career readiness standard on the SAT EBRW assessment.
The problems I have described extend far beyond Wheat Ridge. High school reading remediation is a critical issue all across the nation’s 36th largest school district.
On the 2017 CMAS assessment, 641 Jeffco ninth graders were more than 2 SDs below the meets/exceeds cut score, 862 were between one and two SDs below, and 1,410 were less than one below. These 2,913 students with reading deficits represented 45% of the total number of Jeffco ninth graders Note also that only 33% of Jeffco students qualify for free and reduced lunch — you cannot blame all these reading deficits on poverty.
Across all four high school grades, this suggests there are almost 12,000 students in Jeffco whose lack of proficient reading skill is setting them up for a lifetime of struggle in our intensely competitive knowledge-based economy. This is consistent with results from the ACT Reading assessment that was given to every Jeffco 11th grader until Colorado switched to the SAT in 2017 (which does not break out reading separately). Year after year, over half of Jeffco 11th graders fell short of the ACT’s college and career ready standard for reading ability.
At their December 1st meeting, Dr. Glass told the Jeffco Board of Education, “the funding problems at Wheat Ridge have been resolved.”
Nothing could be further from the truth.
When other business people — many of whom have been through difficult but successful turnarounds — ask me why K-12 results continue to stagnate, despite the billions of taxpayer dollars we spend on our public schools, I tell them about my experiences at Wheat Ridge and note that they are sadly far from unique in American education today.
Tom Coyne is a former public company CEO, and former Chair of the Wheat Ridge High School Accountability Committee