Linda Darling-Hammond

CREDIT: LPI

Linda Darling-Hammond

When the new Smarter Balanced test scores (part of California’s new assessment system known as CAASPP) came out last month, Californians engaged in what is now a time-honored routine — noting that scores inched up but large equity gaps remain.

Differentials by race, ethnicity, income, language background and disability status have become a common feature of the landscape, like our dry riverbeds and brush fires in the summertime — worrisome but not really viewed as solvable.

Thus, we typically have a “Home Alone” moment when we register wide-eyed shock as test score gaps are announced and then go about our business for another year without a purposeful analysis and determination to change the status quo.

Leib Sutcher

Learning Policy Institute

Leib Sutcher

We believe it is important to dive into the data and take stock of what we can learn in order to identify the implications for educational improvement in the state and think carefully about next steps. This takes some doing, however. The state currently reports trends in terms of the proportion of students who “meet the standard,” a cut point set for each grade level, but does not focus on trends in the actual scores the students receive. (These scores are on a scale from 2000 to 3000 on the tests.) Users must go into the data set to calculate these trends themselves across test areas, years, and student subgroups.

Analyzing these trends in actual scores is especially important for understanding the performance of groups scoring well below or well above the “standards met” cut point used for reporting the data.

The Learning Policy Institute looked more closely at these data, tallying scale scores for student subgroups for each year while weighting scores from each grade level and year by the number of test takers to measure how much “growth” this year’s test scores showed compared to last year. We found some surprises offering both good news and bad news — and some clear indicators of where the state needs to dig in and work harder.

First the bad news: As the graphs below show, students with disabilities score low and show little growth. And many historically underserved racial/ethnic groups score well below white and Asian students and are gaining less rapidly, which contributes to a widening achievement gap.  This includes African-American students, where the news is the worst, followed by Native American and Pacific Islander students.

Graph showing growth in scale score by subgroup in English.

Source: Learning Policy Institute

Note: Average scale scores are calculated by taking a weighted average of mean scales scores at each grade level according to the number of students with test scores. The numbers on the growth axis refer to the average gain in the number of scale score points for that group between 2015 and 2016.

And all of the news is worse in mathematics than in English language arts where both average scores and average growth are lower overall and for each group.

Graph showing growth in scale score by subgroup in mathematics.

Source: Learning Policy Institute

Note: Average scale scores are calculated by taking a weighted average of mean scales scores at each grade level according to the number of students with test scores. The numbers on the growth axis refer to the average gain in the number of scale score points for that group between 2015 and 2016.

Gains for Latino students, however, were much stronger than for other  groups, and in English language arts they slightly surpassed the gains for white students. The gap also narrowed for economically disadvantaged students in relation to non-economically disadvantaged students in English language arts, although a substantial gap remains. Filipino students had by far the largest gains of any group and outscored white students in both English language arts and mathematics. Asian students had the highest scores in both subjects and also experienced very strong gains.

These gains for many Latino, Filipino and Asian students were part of an untold success story for California, which is the progress of most English learners over the course of their school careers.

While students currently classified as English learners score low in both English language arts and math, this is in large part because students are no longer counted in the English learner (EL) group as soon as they become reclassified as proficient. (Unlike the system for federal accountability reporting, in this state report, students do not remain in the EL category for several years after they are reclassified.) This catch-22 makes it impossible for the EL group as a whole to show much gain in achievement, even when individual learners are making strong progress. More importantly, as students progress through California schools, the vast majority are being reclassified as English proficient.

When the current 11th graders were in 3rd grade, for example, nearly 150,000 students were classified as English learners. Eight years later, by 11th grade, only about 47,000 students were so classified. By 11th grade, over 125,000 students have been reclassified as English proficient (as compared to only about 60,000 students when this cohort was in 3rd grade.) Even more impressive is the fact that students who are reclassified as fluent English proficient outperform native English speakers in both English language arts and math. Other students from homes where English is not the first language but who were initially fluent English proficient (IFEP) when they entered the school system outperform all other students and continue to make gains at a very high rate. (The IFEP students are likely to be students from more educated and affluent families.)

These data are a testament to the capacity of California schools to help many students acquire the English language and achieve at high levels. The state must still be concerned, of course, about the high school students who are not yet proficient in English — about ¼ of those who were ever English learners. Some of these are recent immigrants, but others have been struggling to become proficient over a longer period of time. The state should monitor growth for these long-time ELs in particular and flag districts where progress is not being made, so that they can get assistance to improve program design and instruction.

Graph showing English learner growth in English/Language Arts scale score by language fluency.

Source: Learning Policy Institute

Note: Average scale scores are calculated by taking a weighted average of mean scales scores at each grade level according to the number of students with test scores. The numbers on the growth axis refer to the average gain in the number of scale score points for that group between 2015 and 2016.

Graph showing English learner growth in math scale score by language fluency.

Source: Learning Policy Institute

Note: Average scale scores are calculated by taking a weighted average of mean scales scores at each grade level according to the number of students with test scores. The numbers on the growth axis refer to the average gain in the number of scale score points for that group between 2015 and 2016.

But what about the areas where California is failing to close the achievement gap and where students are falling further and further behind? Here we have to take a long hard look at the issues that must be put more forcefully on the policy agenda in California.

  • First, the status of special education must be improved. California has lagged dreadfully behind other states in the quality of services and outcomes, with fewer students in mainstream classes, lower achievement and graduation rates, and huge numbers of teachers underprepared relative to other states and to California’s own recently low standards. Because of severe shortages, half of the new entrants to special education teaching in recent years had not completed even a short program of preparation, lacking student teaching and coursework needed to help them do their jobs well. That special education is a five alarm fire in California was acknowledged by the recent state Task Force on Special Education, which laid out a compelling vision for reform. And the California Teacher Credentialing Commission has started work on improving preparation for both general education teachers and specialists who teach these students. But much work needs to be done to realize that vision, beginning with service scholarships and program investments to help candidates become well-trained for their positions, and continuing with new approaches to funding and organizing special education programs.
  • Shortages of teachers are also part of the problem California faces with respect to mathematics achievement, especially in the districts with concentrations of students in poverty that have still not caught up to the spending and salary levels of wealthier districts and have a much harder time competing in the labor market for teachers. Severe shortages of teachers in math, science and special education hit many districts last year and have grown worse in many districts this year. Without inducements like the APLE loans, Cal T grants, Governors fellowships and National Board stipends used in past eras of shortage, the California Teacher Credentialing Commission found itself issuing more than 7,700 substandard credentials and permits last year to fill shortages — more than 1/3 of the total credentials issued — and will be called on to issue even more this year. Each of these actions means putting an underprepared teacher or substitute who lacks knowledge of content and/or teaching strategies in front of students. As has been the case historically, most of these students are low-income students of color in high-need districts, including those falling further and further behind.
  • Finally, we need to acknowledge that for all of the progress associated with the very progressive Local Control Funding Formula — including efforts to direct funds to the schools where those with greater needs are served — most districts have still not recovered the revenues they lost during the years of severe budget cuts. Furthermore, the urban school districts serving most of the state’s African American student population have the greatest ground to make up, as most spent below the state average under the state’s old funding formula even though they served students with the greatest need. These districts need to rebuild staffing, invest in curriculum, technology and professional development and turn around a curriculum that was focused on low-level skills under the old tests that were tied to hard-edged sanctions. (Many Latino students are also served in these districts, but Latinos are more widely spread across a variety of districts in the state.) In addition, unequal allocations of resources within many districts have historically disadvantaged schools serving African-American students — and this also needs attention as we take seriously the idea that Black Minds Matter, along with those of all other groups of students.
Students taking test

Alison Yin for EdSource

Fifth graders take a reading test on Chrome computers at Redwood Heights Elementary School in Oakland.

In sum, a closer look at this year’s CAASPP scores reveals some surprising areas of progress, including above average achievement for reclassified English learners — who represent the vast majority of those who began as English learners and who now outscore English-only students in English language arts and mathematics — as well as a narrowing socioeconomic test-score gaps in English language arts. The data also illuminate substantial and widening gaps for other groups, with African-American students and students with disabilities doing especially poorly. Finally, most students are making less progress in mathematics than in English language arts.

To solve these issues, California must improve the quality of special education services; eliminate teacher shortages in special education and math, as well as other fields; and increase the adequacy and equity of school funding. As with our drought, the longer it takes to put a full range of solutions in place, the more achievement gaps will create lasting damage. This year we hope not to let the test score headlines come and go without taking stock that can lead to action.

•••

Linda Darling-Hammond is president and CEO of the Learning Policy Institute. Leib Sutcher is research associate and a member of LPI’s Educator Quality Team.

The opinions expressed in this commentary represent those of the author. EdSource welcomes commentaries representing diverse points of view. If you would like to submit a commentary, please review our guidelines and contact us.

SHARE ARTICLE

Comments (3)

Leave a Comment

Your email address will not be published. Required fields are marked *

Comments Policy

The goal of the comments section on EdSource is to facilitate thoughtful conversation about content published on our website. Click here for EdSource's Comments Policy.

  1. el 3 months ago3 months ago

    As a layperson, I find it difficult to understand how to proceed in terms of the issues related to students with disabilities cited by this data. Doug McRae brings up some interesting issues as well, and I end up wondering if this just isn't too big a bucket to solve for in one go. Some of these kids are autistic, some have profound cognitive impairment, some are physically limited, some are deaf or blind, some … Read More

    As a layperson, I find it difficult to understand how to proceed in terms of the issues related to students with disabilities cited by this data. Doug McRae brings up some interesting issues as well, and I end up wondering if this just isn’t too big a bucket to solve for in one go.

    Some of these kids are autistic, some have profound cognitive impairment, some are physically limited, some are deaf or blind, some have behavioral issues… there isn’t, to my mind a one-size-fits-all solution for this grab bag group of kids that “aren’t like the others.” A child who is deaf should be able to hit the top of the scale; a child with severe cognitive impairment maybe cannot regardless of the quality of instruction. Some children may be learning valuable skills and be unable to express them and their significant talents on exams such as we give, for example fantastic at pattern recognition but unable to find nuance out of a literature passage.

    There is a tension between breaking out subgroups to sizes where they are large enough to be statistically meaningful and easy to populate and yet small enough to have common concerns.

    The special education shortage of teachers is real, and a bit self-fulfilling, in that I’ve happened on a few conversations where people who were thinking about going for the special ed training were actively advised not to because it would almost ensure that their districts would forcibly assign them special ed rather than their main teaching interest. I hear many reports that the paperwork in special ed, while well-intended, is becoming soul-crushing to people who honestly want to help kids and spend time helping the kids. This paperwork may be driving people away more than the need for training or the lack of funding. No question that the paperwork evolved to stop bad actors from depriving kids of resources, but we might be ready to consider if it is also stopping good actors from giving kids everything they need, and if there are changes that would be beneficial.

    I’m happy with the progressive aims of LCFF, but in practice, as is mentioned, we are still way underfunding schools, and the formula again assumes the worst of all school districts, that before LCFF that they deprived high need students. Some certainly did, but a better measure than “before and after” is “what are you doing now.” A low base grant and general fund hurts high need students too.

    Teachers need to be compensated enough to live their lives and pay their bills without worry. We also need to create stability, so that teachers know that if they get the training, and if they apply themselves and do a good job, that they will have a stable career. The whiplash cuts and funding restorations are extremely damaging to recruitment and continuity. We can’t pay less and have less job security and require more training and expect that’s going to make teaching attractive as a life choice.

  2. Private Guide 4 months ago4 months ago

    Great and useful article!
    It has practical value to future employers.
    It has long been observed that Filipino students are most in demand in the field of internet business and tourism.
    This study only confirms statistics from the Internet.
    Thanks to the authors

  3. Doug McRae 4 months ago4 months ago

    Three observations on this commentary: First, the commentary assumes Smarter Balanced tests do an adequate job of measuring students adversely affected by the so-called 'equity gaps.' Material in the Smarter Balanced submission for federal peer review last June calls this assumption into question -- the SBAC submission acknowledged that the 2015 SBAC tests had "gaps in item coverage at the low end of the performance spectrum" that clearly compromised the accuracy of results for groups affected … Read More

    Three observations on this commentary:

    First, the commentary assumes Smarter Balanced tests do an adequate job of measuring students adversely affected by the so-called ‘equity gaps.’ Material in the Smarter Balanced submission for federal peer review last June calls this assumption into question — the SBAC submission acknowledged that the 2015 SBAC tests had “gaps in item coverage at the low end of the performance spectrum” that clearly compromised the accuracy of results for groups affected by the equity gaps. In other words, SBAC in 2015 did not have a sufficient number of test questions in their computer-adaptive item banks to provide for adequate score estimates for low scoring students. No evidence has been released (as far as I know) that this flaw in the 2015 tests was corrected in time for the 2016 Smarter Balanced tests. The conclusion to be drawn from this information is that no analysis of SBAC scores, either 2015 or 2016, either scale scores or percent proficient scores, provides a good basis for measuring either the status or year-to-year changes for equity gaps in California.

    Second, in the section on English Learners, the commentary states ” . . . as students progress through CA schools, the vast majority are being reclassified as English proficient.” Is there a longitudinal study that can be cited to support this statement? The next few lines of the commentary go on to cite data for current 11th graders (with 47,000 classified at ELs) compared to 150,000 classified of EL’s when they were in 3rd grade. But, that analysis does not take into account that many EL’s in 3rd grade may not have been enrolled 8 years later in CA public schools as 11th graders [i.e., no longer living in CA, attending non-public schools, in-school but not classified as an 11th grader, dropped out]. There may be evidence to support the quoted material above, but it is not cited.

    Finally, the commentary focus on Smarter Balanced scores for Students with Disabilities is well placed. The data in the graphs clearly show both the status and gains for SWDs taking Smarter Balanced tests should be a major concern. CA needs to take a long hard look not only whether curriculum and instruction is adequate for these students, but first we need to ask the question whether the Smarter Balanced tests are adequate measures for these students, particularly with the Smarter Balanced acknowledged gap in item coverage for the low end of the performance spectrum noted above.