Tag Archives: Reliability

Special Issue of “Educational Researcher” (Paper #8 of 9, Part I): A More Research-Based Assessment of VAMs’ Potentials

Posted on by
2

Recall that the peer-reviewed journal Educational Researcher (ER) – published a “Special Issue” including nine articles examining value-added measures (VAMs). I have reviewed the next of nine articles (#8 of 9), which is actually a commentary titled “Can Value-Added Add Value to Teacher Evaluation?” This commentary is authored by Linda Darling-Hammond – Professor of Education, Emeritus, at Stanford University.

Like with the last commentary reviewed here, Darling-Hammond reviews some of the key points taken from the five feature articles in the aforementioned “Special Issue.” More specifically, though, Darling-Hammond “reflect[s] on [these five] articles’ findings in light of other work in this field, and [she] offer[s her own] thoughts about whether and how VAMs may add value to teacher evaluation” (p. 132).

She starts her commentary with VAMs “in theory,” in that VAMs COULD accurately identify teachers’ contributions to student learning and achievement IF (and this is a big IF) the following three conditions were met: (1) “student learning is well-measured by tests that reflect valuable learning and the actual achievement of individual students along a vertical scale representing the full range of possible achievement measures in equal interval units” (2) “students are randomly assigned to teachers within and across schools—or, conceptualized another way, the learning conditions and traits of the group of students assigned to one teacher do not vary substantially from those assigned to another;” and (3) “individual teachers are the only contributors to students’ learning over the period of time used for measuring gains” (p. 132).

None of things are actual true (or near to true, nor will they likely ever be true) in educational practice, however. Hence, the errors we continue to observe that continue to prevent VAM use for their intended utilities, even with the sophisticated statistics meant to mitigate errors and account for the above-mentioned, let’s call them, “less than ideal” conditions.

Other pervasive and perpetual issues surrounding VAMs as highlighted by Darling-Hammond, per each of the three categories above, pertain to (1) the tests used to measure value-added is that the tests are very narrow, focus on lower level skills, and are manipulable. These tests in their current form cannot effectively measure the learning gains of a large share of students who are above or below grade level given a lack of sufficient coverage and stretch. As per Haertel (2013, as cited in Darling-Hammond’s commentary), this “translates into bias against those teachers working with the lowest-performing or the highest-performing classes’…and “those who teach in tracked school settings.” It is also important to note here that the new tests created by the Partnership for Assessing Readiness for College and Careers (PARCC) and Smarter Balanced, multistate consortia “will not remedy this problem…Even though they will report students’ scores on a vertical scale, they will not be able to measure accurately the achievement or learning of students who started out below or above grade level” (p.133).

With respect to (2) above, on the equivalence (or rather non-equivalence) of groups of student across teachers classrooms who are the ones whose VAM scores are relativistically compared, the main issue here is that “the U.S. education system is the one of most segregated and unequal in the industrialized world…[likewise]…[t]he country’s extraordinarily high rates of childhood poverty, homelessness, and food insecurity are not randomly distributed across communities…[Add] the extensive practice of tracking to the mix, and it is clear that the assumption of equivalence among classrooms is far from reality” (p. 133). Whether sophisticated statistics can control for all of this variation is one of most debated issues surrounding VAMs and their levels of outcome bias, accordingly.

And as per (3) above, “we know from decades of educational research that many things matter for student achievement aside from the individual teacher a student has at a moment in time for a given subject area. A partial list includes the following [that are also supposed to be statistically controlled for in most VAMs, but are also clearly not controlled for effectively enough, if even possible]: (a) school factors such as class sizes, curriculum choices, instructional time, availability of specialists, tutors, books, computers, science labs, and other resources; (b) prior teachers and schooling, as well as other current teachers—and the opportunities for professional learning and collaborative planning among them; (c) peer culture and achievement; (d) differential summer learning gains and losses; (e) home factors, such as parents’ ability to help with homework, food and housing security, and physical and mental support or abuse; and (e) individual student needs, health, and attendance” (p. 133).

“Given all of these influences on [student] learning [and achievement], it is not surprising that variation among teachers accounts for only a tiny share of variation in achievement, typically estimated at under 10%” (see, for example, highlights from the American Statistical Association’s (ASA’s) Position Statement on VAMs here). “Suffice it to say [these issues]…pose considerable challenges to deriving accurate estimates of teacher effects…[A]s the ASA suggests, these challenges may have unintended negative effects on overall educational quality” (p. 133). “Most worrisome [for example] are [the] studies suggesting that teachers’ ratings are heavily influenced [i.e., biased] by the students they teach even after statistical models have tried to control for these influences” (p. 135).

Other “considerable challenges” include: VAM output are grossly unstable given the swings and variations observed in teacher classifications across time, and VAM output are “notoriously imprecise” (p. 133) given the other errors observed as caused, for example, by varying class sizes (e.g., Sean Corcoran (2010) documented with New York City data that the “true” effectiveness of a teacher ranked in the 43rd percentile could have had a range of possible scores from the 15th to the 71st percentile, qualifying as “below average,” “average,” or close to “above average). In addition, practitioners including administrators and teachers are skeptical of these systems, and their (appropriate) skepticisms are impacting the extent to which they use and value their value-added data, noting that they value their observational data (and the professional discussions surrounding them) much more. Also important is that another likely unintended effect exists (i.e., citing Susan Moore Johnson’s essay here) when statisticians’ efforts to parse out learning to calculate individual teachers’ value-added causes “teachers to hunker down and focus only on their own students, rather than working collegially to address student needs and solve collective problems” (p. 134). Related, “the technology of VAM ranks teachers against each other relative to the gains they appear to produce for students, [hence] one teacher’s gain is another’s loss, thus creating disincentives for collaborative work” (p. 135). This is what Susan Moore Johnson termed the egg-crate model, or rather the egg-crate effects.

Darling-Hammond’s conclusions are that VAMs have “been prematurely thrust into policy contexts that have made it more the subject of advocacy than of careful analysis that shapes its use. There is [good] reason to be skeptical that the current prescriptions for using VAMs can ever succeed in measuring teaching contributions well (p. 135).

Darling-Hammond also “adds value” in one whole section (highlighted in another post forthcoming here), offering a very sound set of solutions, using VAMs for teacher evaluations or not. Given it’s rare in this area of research we can focus on actual solutions, this section is a must read. If you don’t want to wait for the next post, read Darling-Hammond’s “Modest Proposal” (p. 135-136) within her larger article here.

In the end, Darling-Hammond writes that, “Trying to fix VAMs is rather like pushing on a balloon: The effort to correct one problem often creates another one that pops out somewhere else” (p. 135).

*****

If interested, see the Review of Article #1 – the introduction to the special issue here; see the Review of Article #2 – on VAMs’ measurement errors, issues with retroactive revisions, and (more) problems with using standardized tests in VAMs here; see the Review of Article #3 – on VAMs’ potentials here; see the Review of Article #4 – on observational systems’ potentials here; see the Review of Article #5 – on teachers’ perceptions of observations and student growth here; see the Review of Article (Essay) #6 – on VAMs as tools for “egg-crate” schools here; and see the Review of Article (Commentary) #7 – on VAMs situated in their appropriate ecologies here.

Article #8, Part I Reference: Darling-Hammond, L. (2015). Can value-added add value to teacher evaluation? Educational Researcher, 44(2), 132-137. doi:10.3102/0013189X15575346

“Arbitrary and Capricious:” Sheri Lederman Wins Lawsuit in NY’s State Supreme Court

Posted on by
4

Recall the New York lawsuit pertaining to Long Island teacher Sheri Lederman? She just won in New York’s State Supreme court, and boy did she win big, also for the cause!

Sheri is a teacher, who by all accounts other than her 2013-2014 “ineffective” growth score of a 1/20, is a terrific 4th grade, 18-year veteran teacher. However, after receiving her “ineffective” growth rating and score, she along with her attorney and husband Bruce Lederman, sued the state of New York to challenge the state’s growth-based teacher evaluation system and Sheri’s individual score. See prior posts about Sheri’s case here, here, here and here.

The more specific goal of her case was to seek a judgment: (1) setting aside or vacating Sheri’s individual growth score and rating her as “ineffective,” and (2) declare that the New York endorsed and implemented growth measures in use was/is “arbitrary and capricious.” The “overall gist” was that Sheri contended that the system unfairly penalized teachers whose students consistently scored well and could not demonstrated growth upwards (e.g., teachers of gifted or other high achieving students). This concern/complaint is common elsewhere.

As per a State Supreme Court ruling, just released today as written by Acting Supreme Court Justice Judge Roger McDonough (May 10, 2016), and at 15 pages in length and available in full here, Sheri won her case. She won it against John King — the then New York State Education Department Commissioner and the now US Secretary of Education (who recently replaced Arne Duncan as US Secretary of Education). The Court concluded that Sheri (her husband, her team of experts, and other witnesses) effectively established that her growth score and rating for 2013-2014 was “arbitrary and capricious,” with “arbitrary and capricious” being defined as actions “taken without sound basis in reason or regard to the facts.”

More specifically, the Court’s conclusion was founded upon: (1) the convincing and detailed evidence of VAM bias against teachers at both ends of the spectrum (e.g. those with high-performing students or those with low-performing students); (2) the disproportionate effect of petitioner’s small class size and relatively large percentage of high-performing students; (3) the functional inability of high-performing students to demonstrate growth akin to lower-performing students; (4) the wholly unexplained swing in petitioner’s growth score from 14 [i.e., her growth score the year prior] to 1, despite the presence of statistically similar scoring students in her respective classes; and, most tellingly, (5) the strict imposition of rating constraints in the form of a “bell curve” that places teachers in four categories via pre-determined percentages regardless of whether the performance of students dramatically rose or dramatically fell from the previous year.”

As per an email I received earlier today from Bruce (i.e., Sheri’s husband/attorney who prosecuted her case), the Court otherwise “declined to make an overall ruling on the [New York growth] rating system in general because of new regulations in effect” [e.g., that the state’s growth model is currently under review]…[Nontheless, t]he decision should qualify as persuasive authority for other teachers challenging growth scores throughout the County [and Country]. [In addition, the] Court carefully recite[d] all our expert affidavits [i.e., from Professors Darling-Hammond, Pallas, Amrein-Beardsley, Sean Corcoran and Jesse Rothstein as well as Drs. Burris and Lindell].” Noted as well were the “absence of any meaningful’ challenge to [Sheri’s] experts’ conclusions, especially about the dramatic swings noticed between her, and potentially others’ scores, and the other ‘litany of expert affidavits submitted on [Sheris’] behalf].”

“It is clear that the evidence all of these amazing experts presented was a key factor in winning this case since the Judge repeatedly said both in Court and in the decision that we have a “high burden” to meet in this case.” [In addition,] [t]he Court wrote that the court “does not lightly enter into a critical analysis of this matter … [and] is constrained on this record, to conclude that [the] petitioner [i.e., Sheri] has met her high burden.”

To Bruce’s/our knowledge, this is the first time a judge has set aside an individual teacher’s VAM rating based upon such a presentation in court.

Thanks to all who helped in this endeavor. Onward!

Alabama’s “New” Accountability System: Part II

Posted on by
Reply

In a prior post, about whether the state of “Alabama is the New, New Mexico,” I wrote about a draft bill in Alabama to be called the Rewarding Advancement in Instruction and Student Excellence (RAISE) Act of 2016. This has since been renamed the Preparing and Rewarding Educational Professionals (PREP) Bill (to be Act) of 2016. The bill was introduced by its sponsoring Republican Senator Marsh last Tuesday, and its public hearing is scheduled for tomorrow. I review the bill below, and attach it here for others who are interested in reading it in full.

First, the bill is to “provide a procedure for observing and evaluating teachers, principals, and assistant principals on performance and student achievement…[using student growth]…to isolate the effect and impact of a teacher on student learning, controlling for pre-existing characteristics of a student including, but not limited to, prior achievement.” Student growth is still one of the bill’s key components, with growth set at a 25% weight, and this is still written into this bill regardless of the fact that the new federal Elementary Student Success Act (ESSA) no longer requires teacher-level growth as a component of states’ educational reform legislation. In other words, states are no longer required to do this, but apparently the state/Senator Marsh still wants to move forward in this regard, regardless (and regardless of the research evidence). The student growth model is to be selected by October 1, 2016. On this my offer (as per my prior post) still stands. I would  be more than happy to help the state negotiate this contract, pro bono, and much more wisely than so many other states and districts have negotiated similar contracts thus far (e.g., without asking for empirical evidence as a continuous contractual deliverable).

Second, and related, nothing is written about the ongoing research and evaluation of the state system, that is absolutely necessary in order to ensure the system is working as intended, especially before any types of consequential decisions are to be made (e.g., school bonuses, teachers’ denial of tenure, teacher termination, teacher termination due to a reduction in force). All that is mentioned is that things like stakeholder perceptions, general outcomes, and local compliance with the state will be monitored. Without evidence in hand, in advance and preferably as externally vetted and validated, the state will very likely be setting itself up for some legal trouble.

Third, to measure growth the state is set to use student performance data on state tests, as well as data derived via the ACT Aspire examination, American College Test (ACT), and “any number of measures from the department developed list of preapproved options for governing boards to utilize to measure student achievement growth.” As mentioned in my prior post about Alabama (linked to again here), this is precisely what has gotten the whole state of New Mexico wrapped up in, and quasi-losing their ongoing lawsuit. While providing districts with menus of off-the-shelf and other assessment options might make sense to policymakers, any self respecting researcher should know why this is entirely inappropriate. To read more about this, the best research study explaining why doing just this will set any state up for lawsuits comes from Brown University’s John Papay in his highly esteemed and highly cited “Different tests, different answers: The stability of teacher value-added estimates across outcome measures” article. The title of this research article alone should explain enough why simply positioning and offering up such tests in such casual ways makes way for legal recourse.

Fourth, the bill is to increase the number of years it will take Alabama teachers to earn tenure, requiring that teachers teach for at least five years, of which at least three  consecutive years of “satisfies expectations,” “exceeds expectations,” or “significantly exceeds expectations” are demonstrated via the state’s evaluation system prior to earning tenure. Clearly the state does not understand the current issues with value-added/growth levels of reliability, or consistency, or lack thereof, that are altogether preventing such consistent classifications of teachers over time. Inversely, what is consistently evident across all growth models is that estimates are very inconsistent from year to year, which will likely thwart what the bill has written into it here, as such a theoretically simple proposition. For example, the common statistic still cited in this regard is that a  teacher classified as “adding value” has a 25% to 50% chance of being classified as “subtracting value” the following year(s), and vice versa. This sometimes makes the probability of a teacher being consistently identified as (in)effective, from year-to-year, no different than the flip of a coin, and this is true when there are at least three years of data (which is standard practice and is also written into this bill as a minimum requirement).

Unless the state plans on “artificially conflating” scores, by manufacturing and forcing the oft-unreliable growth data to fit or correlate with teachers’ observational data (two observations per year are to be required), and/or survey data (student surveys are to be used for teachers of students in grades three and above), such consistency is thus far impossible unless deliberately manipulated (see a recent post here about how “artificial conflation” is one of the fundamental and critical points of litigation in another lawsuit in Houston). Related, this bill is also to allow a governing board to evaluate a tenured teacher as per his/her similar classifications every other year, and is to subject a tenured teacher who received a rating of below or significantly below expectations for two consecutive evaluations to personnel action.

Fifth and finally, the bill is also to use an allocated $5,000,000 to recruit teachers, $3,000,000 to mentor teachers, and $10,000,000 to reward the top 10% of schools in the state as per their demonstrated performance. The latter of these is the most consequential as per the state’s use of its planned growth data at the school level (see other teacher-level consequences to be attached to growth output above); hence, the comments about needing empirical evidence to justify such allocations prior to the distributions of such monies is important to underscore again. Likewise, given levels of bias in value-added/growth output are worse at the state versus teacher levels, I would also caution the state against rewarding schools, again in this regard, for what might not really be schools’ causal impacts on student growth over time, after all. See, for example, here, here, and here.

As I also mentioned in my prior post on Alabama, for those of you who have access to educational leaders there, do send them this post too, so they might be a bit more proactive, and appropriately more careful and cautious, before going down what continues to demonstrate itself as a poor educational policy path. While I do embrace my professional responsibility as a public scholar to be called to court to testify about all of this when such high-stakes consequences are ultimately, yet inappropriately based upon invalid inferences, I’d much rather be proactive in this regard and save states and states’ taxpayers their time and money, respectively.

Tennessee’s Trout/Taylor Value-Added Lawsuit Dismissed

Posted on by
1

As you may recall, one of 15 important lawsuits pertaining to teacher value-added estimates across the nation (Florida n=2, Louisiana n=1, Nevada n=1, New Mexico n=4, New York n=3, Tennessee n=3, and Texas n=1 – see more information here) was situated in Knox County, Tennessee.

Filed in February of 2015, with legal support provided by the Tennessee Education Association (TEA), Knox County teacher Lisa Trout and Mark Taylor charged that they were denied monetary bonuses after their Tennessee Value-Added Assessment System (TVAAS — the original Education Value-Added Assessment System (EVAAS)) teacher-level value-added scores were miscalculated. This lawsuit was also to contest the reasonableness, rationality, and arbitrariness of the TVAAS system, as per its intended and actual uses in this case, but also in Tennessee writ large. On this case, Jesse Rothstein (University of California – Berkeley) and I were serving as the Plaintiffs’ expert witnesses.

Unfortunately, however, last week (February 17, 2016) the Plaintiffs’ team received a Court order written by U.S. District Judge Harry S. Mattice Jr. dismissing their claims. While the Court had substantial questions about the reliability and validity of the TVAAS, the Court determined that the State satisfied the very low threshold of the “rational basis test,” at legal issue. I should note here, however, that all of the evidence that the lawyers for the Plaintiffs collected via their “extensive discovery,” including the affidavits both Jesse and I submitted on Plaintiffs’ behalves, were unfortunately not considered in Judge Mattice’s motion to dismiss. This, perhaps, makes sense given some of the assertions made by the Court, forthcoming.

Ultimately, the Court found that the TVAAS-based, teacher-level value-added policy at issue was “rationally related to a legitimate government interest.” As per the Court order itself, Judge Mattice wrote that “While the court expresses no opinion as to whether the Tennessee Legislature has enacted sound public policy, it finds that the use of TVAAS as a means to measure teacher efficacy survives minimal constitutional scrutiny. If this policy proves to be unworkable in practice, plaintiffs are not to be vindicated by judicial intervention but rather by democratic process.”

Otherwise, as per an article in the Knoxville News Sentinel, Judge Mattice was “not unsympathetic to the teachers’ claims,” for example, given the TVAAS measures “student growth — not teacher performance — using an algorithm that is not fail proof.” He inversely noted, however, in the Court order that the “TVAAS algorithms have been validated for their accuracy in measuring a teacher’s effect on student growth,” even if minimal. He also wrote that the test scores used in the TVAAS (and other models) “need not be validated for measuring teacher effectiveness merely because they are used as an input in a validated statistical model that measures teacher effectiveness.” This is, unfortunately, untrue. Nonetheless, he continued to write that even though the rational basis test “might be a blunt tool, a rational policymaker could conclude that TVAAS is ‘capable of measuring some marginal impact that teachers can have on their own students…[and t]his is all the Constitution requires.”

In the end, Judge Mattice concluded in the Court order that, overall, “It bears repeating that Plaintiff’s concerns about the statistical imprecision of TVAAS are not unfounded. In addressing Plaintiffs’ constitutional claims, however, the Court’s role is extremely limited. The judiciary is not empowered to second-guess the wisdom of the Tennessee legislature’s approach to solving the problems facing public education, but rather must determine whether the policy at issue is rationally related to a legitimate government interest.”

It is too early to know whether the prosecution team will appeal, although Judge Mattice dismissed the federal constitutional claims within the lawsuit “with prejudice.” As per an article in the Knoxville News Sentinel, this means that “it cannot be resurrected with new facts or legal claims or in another court. His decision can be appealed, though, to the 6th Circuit U.S. Court of Appeals.”

A Retired Massachusetts Principal on her Teachers’ “Value-Added”

Posted on by
2

A retired Massachusetts principal, named Linda Murdock, posted a post on her blog titled “Murdock’s EduCorner” about her experiences, as a principal, with “value-added,” or more specifically in her state the use of Student Growth Percentile (SGP) scores to estimate said “value-added.” It’s certainly worth reading as one thing I continue to find is that which we continue to find in the research on value-added models (VAMs) is also being realized by practitioners in the schools being required to use value-added output such as these. In this case, for example, while Murdock does not discuss the technical terms we use in the research (e.g., reliability, validity, and bias), she discusses these in pragmatic, real terms (e.g., year-to-year fluctuations, lack of relationship of SGP scores and other indicators of teacher effectiveness, and the extent to which certain sets of students can hinder teachers’ demonstrated growth or value-added, respectively). Hence, do give her post a read here, and also pasted in full below. Do also pay special attention to the bulleted sections in which she discusses these and other issues on a case-by-case basis.

Murdock writes:

At the end of the last school year, I was chatting with two excellent teachers, and our conversation turned to the new state-mandated teacher evaluation system and its use of student “growth scores” (“Student Growth Percentiles” or “SGPs” in Massachusetts) to measure a teacher’s “impact on student learning.”

“Guess we didn’t have much of an impact this year,” said one teacher.

The other teacher added, “It makes you feel about this high,” showing a tiny space between her thumb and forefinger.

Throughout the school, comments were similar — indicating that a major “impact” of the new evaluation system is demoralizing and discouraging teachers. (How do I know, by the way, that these two teachers are excellent? I know because I worked with them as their principal – being in their classrooms, observing and offering feedback, talking to parents and students, and reviewing products demonstrating their students’ learning – all valuable ways of assessing a teacher’s “impact”.)

According to the Massachusetts Department of Elementary and Secondary Education (“DESE”), the new evaluation system’s goals include promoting the “growth and development of leaders and teachers,” and recognizing “excellence in teaching and leading.” The DESE website indicates that the DESE considers a teacher’s median SGP as an appropriate measure of that teacher’s “impact on student learning”:

“ESE has confidence that SGPs are a high quality measure of student growth. While the precision of a median SGP decreases with fewer students, median SGP based on 8-19 students still provides quality information that can be included in making a determination of an educator’s impact on students.”

Given the many concerns about the use of “value-added measurement” tools (such as SGPs) in teacher evaluation, this confidence is difficult to understand, particularly as applied to real teachers in real schools. Considerable research notes the imprecision and variability of these measures as applied to the evaluation of individual teachers. On the other side, experts argue that use of an “imperfect measure” is better than past evaluation methods. Theories aside, I believe that the actual impact of this “measure” on real people in real schools is important.

As a principal, when I first heard of SGPs I was curious. I wondered whether the data would actually filter out other factors affecting student performance, such as learning disabilities, English language proficiency, or behavioral challenges, and I wondered if the data would give me additional information useful in evaluating teachers.

Unfortunately, I found that SGPs did not provide useful information about student growth or learning, and median SGPs were inconsistent and not correlated with teaching skill, at least for the teachers with whom I was working. In two consecutive years of SGP data from our Massachusetts elementary school:

  • One 4th grade teacher had median SGPs of 37 (ELA) and 36 (math) in one year, and 61.5 and 79 the next year. The first year’s class included students with disabilities and the next year’s did not.
  • Two 4th grade teachers who co-teach their combined classes (teaching together, all students, all subjects) had widely differing median SGPs: one teacher had SGPs of 44 (ELA) and 42 (math) in the first year and 40 and 62.5 in the second, while the other teacher had SGPs of 61 and 50 in the first year and 41 and 45 in the second.
  • A 5th grade teacher had median SGPs of 72.5 and 64 for two math classes in the first year, and 48.5, 26, and 57 for three math classes in the following year. The second year’s classes included students with disabilities and English language learners, but the first year’s did not.
  • Another 5th grade teacher had median SGPs of 45 and 43 for two ELA classes in the first year, and 72 and 64 in the second year. The first year’s classes included students with disabilities and students with behavioral challenges while the second year’s classes did not.

As an experienced observer/evaluator, I found that median SGPs did not correlate with teachers’ teaching skills but varied with class composition. Stronger teachers had the same range of SGPs in their classes as teachers with weaker skills, and median SGPs for a new teacher with a less challenging class were higher than median SGPs for a highly skilled veteran teacher with a class that included English language learners.

Furthermore, SGP data did not provide useful information regarding student growth. In analyzing students’ SGPs, I noticed obvious general patterns: students with disabilities had lower SGPs than students without disabilities, English language learners had lower SGPs than students fluent in English, students who had some kind of trauma that year (e.g., parents’ divorce) had lower SGPs, and students with behavioral/social issues had lower SGPs. SGPs were correlated strongly with test performance: in one year, for example, the median ELA SGP for students in the “Advanced” category was 88, compared with 51.5 for “Proficient” students, 19.5 for “Needs Improvement,” and 5 for the “Warning” category.

There were also wide swings in student SGPs, not explainable except perhaps by differences in student performance on particular test days. One student with disabilities had an SGP of 1 in the first year and 71 in the next, while another student had SGPs of 4 in ELA and 94 in math in 4th grade and SGPs of 50 in ELA and 4 in math in 5th grade, both with consistent district test scores.

So how does this “information” impact real people in a real school?  As a principal, I found that it added nothing to what I already knew about the teaching and learning in my school. Using these numbers for teacher evaluation does, however, negatively impact schools: it demoralizes and discourages teachers, and it has the potential to affect class and teacher assignments.

In real schools, student and teacher assignments are not random. Students are grouped for specific purposes, and teachers are assigned classes for particular reasons. Students with disabilities and English language learners are often grouped to allow specialists, such as the speech/language teacher or the ELL teacher, to work more effectively with them. Students with behavioral issues are sometimes placed in special classes, and are often assigned to teachers who work particularly well with them. Leveled classes (AP, honors, remedial), create different student combinations, and teachers are assigned particular classes based on the administrator’s judgment of which teachers will do the best with which classes. For example, I would assign new or struggling teachers less challenging classes so I could work successfully with them on improving their skills.

In the past, when I told a teacher that he/she had a particularly challenging class, because he/she could best work with these students, he/she generally cheerfully accepted the challenge, and felt complimented on his/her skills. Now, that teacher could be concerned about the effect of that class on his/her evaluation. Teachers may be reluctant to teach lower level courses, or to work with English language learners or students with behavioral issues, and administrators may hesitate to assign the most challenging classes to the most skilled teachers.

In short, in my experience, the use of this type of “value-added” measurement provides no useful information and has a negative impact on real teachers and real administrators in real schools. If “data” is not only not useful, but actively harmful, to those who are supposedly benefitting from using it, what is the point? Why is this continuing?

Report on the Stability of Student Growth Percentile (SGP) “Value-Added” Estimates

Posted on by
2

The Student Growth Percentiles (SGPs) model, which is loosely defined by value-added model (VAM) purists as a VAM, uses students’ level(s) of past performance to determine students’ normative growth over time, as compared to his/her peers. “SGPs describe the relative location of a student’s current score compared to the current scores of students with similar score histories” (Castellano & Ho, p. 89). Students are compared to themselves (i.e., students serve as their own controls) over time; therefore, the need to control for other variables (e.g., student demographics) is less necessary, although this is of debate. Nonetheless, the SGP model was developed as a “better” alternative to existing models, with the goal of providing clearer, more accessible, and more understandable results to both internal and external education stakeholders and consumers. For more information about the SGP please see prior posts here and here. See also an original source about the SGP here.

Related, in a study released last week, WestEd researchers conducted an “Analysis of the stability of teacher-level growth scores [derived] from the student growth percentile [SGP] model” in one, large school district in Nevada (n=370 teachers). The key finding they present is that “half or more of the variance in teacher scores from the [SGP] model is due to random or otherwise unstable sources rather than to reliable information that could predict future performance. Even when derived by averaging several years of teacher scores, effectiveness estimates are unlikely to provide a level of reliability desired in scores used for high-stakes decisions, such as tenure or dismissal. Thus, states may want to be cautious in using student growth percentile [SGP] scores for teacher evaluation.”

Most importantly, the evidence in this study should make us (continue to) question the extent to which “the learning of a teacher’s students in one year will [consistently] predict the learning of the teacher’s future students.” This is counter to the claims continuously made by VAM proponents, including folks like Thomas Kane — economics professor from Harvard University who directed the $45 million worth of Measures of Effective Teaching (MET) studies for the Bill & Melinda Gates Foundation. While faint signals of what we call predictive validity might be observed across VAMs, what folks like Kane overlook or avoid is that very often these faint signals do not remain constant over time. Accordingly, the extent to which we can make stable predictions is limited.

Worse is when folks falsely assume that said predictions will remain constant over time, and they make high-stakes decisions about teachers unaware of the lack of stability present, in typically 25-59% of teachers’ value-added (or in this case SGP) scores (estimates vary by study and by analyses using one to three years of data — see, for example, the studies detailed in Appendix A of this report; see also other research on this topic here, here, and here). Nonetheless, researchers in this study found that in mathematics, 50% of the variance in teachers’ value-added scores were attributable to differences among teachers, and the other 50% was random or unstable. In reading, 41% of the variance in teachers’ value-added scores were attributable to differences among teachers, and the other 59% was random or unstable.

In addition, using a 95% confidence interval (which is very common in educational statistics) researchers found that in mathematics, a teacher’s true score would span 48 points, “a margin of error that covers nearly half the 100 point score scale,” whereby “one would be 95 percent confident that the true math score of a teacher who received a score of 50 [would actually fall] between 26 and 74.” For reading, a teacher’s true score would span 44 points, whereby one would be 95 percent confident that the true reading score of a teacher who received a score of 50 would actually fall between 38 and 72. The stability of these scores would increase with three years of data, which has also been found by other researchers on this topic. However, they too have found that such error rates persist to an extent that still prohibits high-stakes decision making.

In more practical terms, what this also means is that a teacher who might be considered highly ineffective might be terminated, even though the following year (s)he could have been observed to be highly effective. Inversely, teachers who are awarded tenure might be observed as ineffective one, two, and/or three years following, not because their true level(s) of effectiveness change, but because of the error in the estimates that causes such instabilities to occur. Hence, examinations of the the stability of such estimates over time provides essential evidence of the validity, and in this case predictive validity, of the interpretations and uses of such scores over time. This is particularly pertinent when high-stakes decisions are to be based on (or in large part on) such scores, especially given some researchers are calling for reliability coefficients of .85 or higher to make such decisions (Haertel, 2013; Wasserman & Bracken, 2003).

In the end, researchers’ overall conclusion is that SGP-derived “growth scores alone may not be sufficiently stable to support high-stakes decisions.” Likewise, relying on the extant research on this topic, the overall conclusion can be broadened in that neither SGP- or VAM-based growth scores may be sufficiently stable to support high-stakes decisions. In other words, it is not just the SGP model that is yielding such issues with stability (or a lack thereof). Again, see the other literature in which researchers situated their findings in Appendix A. See also other similar studies here, here, and here.

Accordingly, those who read this report, and consequently seek to find a better or more stable model that yields more stable estimates, will unfortunately but likely fail in their search.

References:

Castellano, K. E., & Ho, A. D. (2013). A practitioner’s guide to growth models. Washington, DC: Council of Chief State School Officers.

Haertel, E. H. (2013). Reliability and validity of inferences about teachers based on student test scores (14th William H. Angoff Memorial Lecture). Princeton, NJ: Educational Testing Service (ETS).

Lash, A., Makkonen, R., Tran, L., & Huang, M. (2016). Analysis of the stability of teacher-level growth scores [derived] from the student growth percentile [SGP] model. (16–104). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory West.

Wasserman, J. D., & Bracken, B. A. (2003). Psychometric characteristics of assessment procedures. In I. B. Weiner, J. R. Graham, & J. A. Naglieri (Eds.), Handbook of psychology:
Assessment psychology (pp. 43–66). Hoboken, NJ: John Wiley & Sons.

Houston Lawsuit Update, with Summary of Expert Witnesses’ Findings about the EVAAS

Posted on by
1

Recall from a prior post that a set of teachers in the Houston Independent School District (HISD), with the support of the Houston Federation of Teachers (HFT) are taking their district to federal court to fight for their rights as professionals, and how their value-added scores, derived via the Education Value-Added Assessment System (EVAAS), have allegedly violated them. The case, Houston Federation of Teachers, et al. v. Houston ISD, is to officially begin in court early this summer.

More specifically, the teachers are arguing that EVAAS output are inaccurate, the EVAAS is unfair, that teachers are being evaluated via the EVAAS using tests that do not match the curriculum they are to teach, that the EVAAS system fails to control for student-level factors that impact how well teachers perform but that are outside of teachers’ control (e.g., parental effects), that the EVAAS is incomprehensible and hence very difficult if not impossible to actually use to improve upon their instruction (i.e., actionable), and, accordingly, that teachers’ due process rights are being violated because teachers do not have adequate opportunities to change as a results of their EVAAS results.

The EVAAS is the one value-added model (VAM) on which I’ve conducted most of my research, also in this district (see, for example, here, here, here, and here); hence, I along with Jesse Rothstein – Professor of Public Policy and Economics at the University of California – Berkeley, who also conducts extensive research on VAMs – are serving as the expert witnesses in this case.

What was recently released regarding this case is a summary of the contents of our affidavits, as interpreted by authors of the attached “EVAAS Litigation UPdate,” in which the authors declare, with our and others’ research in support, that “Studies Declare EVAAS ‘Flawed, Invalid and Unreliable.” Here are the twelve key highlights, again, as summarized by the authors of this report and re-summarized, by me, below:

  1. Large-scale standardized tests have never been validated for their current uses. In other words, as per my affidavit, “VAM-based information is based upon large-scale achievement tests that have been developed to assess levels of student achievement, but not levels of growth in student achievement over time, and not levels of growth in student achievement over time that can be attributed back to students’ teachers, to capture the teachers’ [purportedly] causal effects on growth in student achievement over time.”
  2. The EVAAS produces different results from another VAM. When, for this case, Rothstein constructed and ran an alternative, albeit sophisticated VAM using data from HISD both times, he found that results “yielded quite different rankings and scores.” This should not happen if these models are indeed yielding indicators of truth, or true levels of teacher effectiveness from which valid interpretations and assertions can be made.
  3. EVAAS scores are highly volatile from one year to the next. Rothstein, when running the actual data, found that while “[a]ll VAMs are volatile…EVAAS growth indexes and effectiveness categorizations are particularly volatile due to the EVAAS model’s failure to adequately account for unaccounted-for variation in classroom achievement.” In addition, volatility is “particularly high in grades 3 and 4, where students have relatively few[er] prior [test] scores available at the time at which the EVAAS scores are first computed.”
  4. EVAAS overstates the precision of teachers’ estimated impacts on growth. As per Rothstein, “This leads EVAAS to too often indicate that teachers are statistically distinguishable from the average…when a correct calculation would indicate that these teachers are not statistically distinguishable from the average.”
  5. Teachers of English Language Learners (ELLs) and “highly mobile” students are substantially less likely to demonstrate added value, as per the EVAAS, and likely most/all other VAMs. This, what we term as “bias,” makes it “impossible to know whether this is because ELL teachers [and teachers of highly mobile students] are, in fact, less effective than non-ELL teachers [and teachers of less mobile students] in HISD, or whether it is because the EVAAS VAM is biased against ELL [and these other] teachers.”
  6. The number of students each teacher teaches (i.e., class size) also biases teachers’ value-added scores. As per Rothstein, “teachers with few linked students—either because they teach small classes or because many of the students in their classes cannot be used for EVAAS calculations—are overwhelmingly [emphasis added] likely to be assigned to the middle effectiveness category under EVAAS (labeled “no detectable difference [from average], and average effectiveness”) than are teachers with more linked students.”
  7. Ceiling effects are certainly an issue. Rothstein found that in some grades and subjects, “teachers whose students have unusually high prior year scores are very unlikely to earn high EVAAS scores, suggesting that ‘ceiling effects‘ in the tests are certainly relevant factors.” While EVAAS and HISD have previously acknowledged such problems with ceiling effects, they apparently believe these effects are being mediated with the new and improved tests recently adopted throughout the state of Texas. Rothstein, however, found that these effects persist even given the new and improved.
  8. There are major validity issues with “artificial conflation.” This is a term I recently coined to represent what is happening in Houston, and elsewhere (e.g., Tennessee), when district leaders (e.g., superintendents) mandate or force principals and other teacher effectiveness appraisers or evaluators, for example, to align their observational ratings of teachers’ effectiveness with value-added scores, with the latter being the “objective measure” around which all else should revolve, or align; hence, the conflation of the one to match the other, even if entirely invalid. As per my affidavit, “[t]o purposefully and systematically endorse the engineering and distortion of the perceptible ‘subjective’ indicator, using the perceptibly ‘objective’ indicator as a keystone of truth and consequence, is more than arbitrary, capricious, and remiss…not to mention in violation of the educational measurement field’s Standards for Educational and Psychological Testing” (American Educational Research Association (AERA), American Psychological Association (APA), National Council on Measurement in Education (NCME), 2014).
  9. Teaching-to-the-test is of perpetual concern. Both Rothstein and I, independently, noted concerns about how “VAM ratings reward teachers who teach to the end-of-year test [more than] equally effective teachers who focus their efforts on other forms of learning that may be more important.”
  10. HISD is not adequately monitoring the EVAAS system. According to HISD, EVAAS modelers keep the details of their model secret, even from them and even though they are paying an estimated $500K per year for district teachers’ EVAAS estimates. “During litigation, HISD has admitted that it has not performed or paid any contractor to perform any type of verification, analysis, or audit of the EVAAS scores. This violates the technical standards for use of VAM that AERA specifies, which provide that if a school district like HISD is going to use VAM, it is responsible for ‘conducting the ongoing evaluation of both intended and unintended consequences’ and that ‘monitoring should be of sufficient scope and extent to provide evidence to document the technical quality of the VAM application and the validity of its use’ (AERA Statement, 2015).
  11. EVAAS lacks transparency. AERA emphasizes the importance of transparency with respect to VAM uses. For example, as per the AERA Council who wrote the aforementioned AERA Statement, “when performance levels are established for the purpose of evaluative decisions, the methods used, as well as the classification accuracy, should be documented and reported” (AERA Statement, 2015). However, and in contrast to meeting AERA’s requirements for transparency, in this district and elsewhere, as per my affidavit, the “EVAAS is still more popularly recognized as the ‘black box’ value-added system.”
  12. Related, teachers lack opportunities to verify their own scores. This part is really interesting. “As part of this litigation, and under a very strict protective order that was negotiated over many months with SAS [i.e., SAS Institute Inc. which markets and delivers its EVAAS system], Dr. Rothstein was allowed to view SAS’ computer program code on a laptop computer in the SAS lawyer’s office in San Francisco, something that certainly no HISD teacher has ever been allowed to do. Even with the access provided to Dr. Rothstein, and even with his expertise and knowledge of value-added modeling, [however] he was still not able to reproduce the EVAAS calculations so that they could be verified.”Dr. Rothstein added, “[t]he complexity and interdependency of EVAAS also presents a barrier to understanding how a teacher’s data translated into her EVAAS score. Each teacher’s EVAAS calculation depends not only on her students, but also on all other students with- in HISD (and, in some grades and years, on all other students in the state), and is computed using a complex series of programs that are the proprietary business secrets of SAS Incorporated. As part of my efforts to assess the validity of EVAAS as a measure of teacher effectiveness, I attempted to reproduce EVAAS calculations. I was unable to reproduce EVAAS, however, as the information provided by HISD about the EVAAS model was far from sufficient.”

Victory in Court: Consequences Attached to VAMs Suspended Throughout New Mexico

Posted on by
4

Great news for New Mexico and New Mexico’s approximately 23,000 teachers, and great news for states and teachers potentially elsewhere, in terms of setting precedent!

Late yesterday, state District Judge David K. Thomson, who presided over the ongoing teacher-evaluation lawsuit in New Mexico, granted a preliminary injunction preventing consequences from being attached to the state’s teacher evaluation data. More specifically, Judge Thomson ruled that the state can proceed with “developing” and “improving” its teacher evaluation system, but the state is not to make any consequential decisions about New Mexico’s teachers using the data the state collects until the state (and/or others external to the state) can evidence to the court during another trial (set for now, for April) that the system is reliable, valid, fair, uniform, and the like.

As you all likely recall, the American Federation of Teachers (AFT), joined by the Albuquerque Teachers Federation (ATF), last year, filed a “Lawsuit in New Mexico Challenging [the] State’s Teacher Evaluation System.” Plaintiffs charged that the state’s teacher evaluation system, imposed on the state in 2012 by the state’s current Public Education Department (PED) Secretary Hanna Skandera (with value-added counting for 50% of teachers’ evaluation scores), is unfair, error-ridden, spurious, harming teachers, and depriving students of high-quality educators, among other claims (see the actual lawsuit here).

Thereafter, one scheduled day of testimonies turned into five in Santa Fe, that ran from the end of September through the beginning of October (each of which I covered here, here, here, here, and here). I served as the expert witness for the plaintiff’s side, along with other witnesses including lawmakers (e.g., a state senator) and educators (e.g., teachers, superintendents) who made various (and very articulate) claims about the state’s teacher evaluation system on the stand. Thomas Kane served as the expert witness for the defendant’s side, along with other witnesses including lawmakers and educators who made counter claims about the system, some of which backfired, unfortunately for the defense, primarily during cross-examination.

See articles released about this ruling this morning in the Santa Fe New Mexican (“Judge suspends penalties linked to state’s teacher eval system”) and the Albuquerque Journal (“Judge curbs PED teacher evaluations).” See also the AFT’s press release, written by AFT President Randi Weingarten, here. Click here for the full 77-page Order written by Judge Thomson (see also, below, five highlights I pulled from this Order).

The journalist of the Santa Fe New Mexican, though, provided the most detailed information about Judge Thomson’s Order, writing, for example, that the “ruling by state District Judge David Thomson focused primarily on the complicated combination of student test scores used to judge teachers. The ruling [therefore] prevents the Public Education Department [PED] from denying teachers licensure advancement or renewal, and it strikes down a requirement that poorly performing teachers be placed on growth plans.” In addition, the Judge noted that “the teacher evaluation system varies from district to district, which goes against a state law calling for a consistent evaluation plan for all educators.”

The PED continues to stand by its teacher evaluation system, calling the court challenge “frivolous” and “a legal PR stunt,” all the while noting that Judge Thomson’s decision “won’t affect how the state conducts its teacher evaluations.” Indeed it will, for now and until the state’s teacher evaluation system is vetted, and validated, and “the court” is “assured” that the system can actually be used to take the “consequential actions” against teachers, “required” by the state’s PED.

Here are some other highlights that I took directly from Judge Thomson’s ruling, capturing what I viewed as his major areas of concern about the state’s system (click here, again, to read Judge Thomson’s full Order):

  • Validation Needed: “The American Statistical Association says ‘estimates from VAM should always be accompanied by measures of precision and a discussion of the assumptions and possible limitations of the model. These limitations are particularly relevant if VAM are used for high stake[s] purposes” (p. 1). These are the measures, assumptions, limitations, and the like that are to be made transparent in this state.
  • Uniformity Required: “New Mexico’s evaluation system is less like a [sound] model than a cafeteria-style evaluation system where the combination of factors, data, and elements are not easily determined and the variance from school district to school district creates conflicts with the [state] statutory mandate” (p. 2)…with the existing statutory framework for teacher evaluations for licensure purposes requiring “that the teacher be evaluated for ‘competency’ against a ‘highly objective uniform statewide standard of evaluation’ to be developed by PED” (p. 4). “It is the term ‘highly objective uniform’ that is the subject matter of this suit” (p. 4), whereby the state and no other “party provided [or could provide] the Court a total calculation of the number of available district-specific plans possible given all the variables” (p. 54). See also the Judge’s points #78-#80 (starting on page 70) for some of the factors that helped to “establish a clear lack of statewide uniformity among teachers” (p. 70).
  • Transparency Missing: “The problem is that it is not easy to pull back the curtain, and the inner workings of the model are not easily understood, translated or made accessible” (p. 2). “Teachers do not find the information transparent or accurate” and “there is no evidence or citation that enables a teacher to verify the data that is the content of their evaluation” (p. 42). In addition, “[g]iven the model’s infancy, there are no real studies to explain or define the [s]tate’s value-added system…[hence, the consequences and decisions]…that are to be made using such system data should be examined and validated prior to making such decisions” (p. 12).
  • Consequences Halted: “Most significant to this Order, [VAMs], in this [s]tate and others, are being used to make consequential decisions…This is where the rubber hits the road [as per]…teacher employment impacts. It is also where, for purposes of this proceeding, the PED departs from the statutory mandate of uniformity requiring an injunction” (p. 9). In addition, it should be noted that indeed “[t]here are adverse consequences to teachers short of termination” (p. 33) including, for example, “a finding of ‘minimally effective’ [that] has an impact on teacher licenses” (p. 41). These, too, are to be halted under this injunction Order.
  • Clarification Required: “[H]ere is what this [O]rder is not: This [O]rder does not stop the PED’s operation, development and improvement of the VAM in this [s]tate, it simply restrains the PED’s ability to take consequential actions…until a trial on the merits is held” (p. 2). In addition, “[a] preliminary injunction differs from a permanent injunction, as does the factors for its issuance…’ The objective of the preliminary injunction is to preserve the status quo [minus the consequences] pending the litigation of the merits. This is quite different from finally determining the cause itself” (p. 74). Hence, “[t]he court is simply enjoining the portion of the evaluation system that has adverse consequences on teachers” (p. 75).

The PED also argued that “an injunction would hurt students because it could leave in place bad teachers.” As per Judge Thomson, “That is also a faulty argument. There is no evidence that temporarily halting consequences due to the errors outlined in this lengthy Opinion more likely results in retention of bad teachers than in the firing of good teachers” (p. 75).

Finally, given my involvement in this lawsuit and given the team with whom I was/am still so fortunate to work (see picture below), including all of those who testified as part of the team and whose testimonies clearly proved critical in Judge Thomson’s final Order, I want to thank everyone for all of their time, energy, and efforts in this case, thus far, on behalf of the educators attempting to (still) do what they love to do — teach and serve students in New Mexico’s public schools.

IMG_0123

Left to right: (1) Stephanie Ly, President of AFT New Mexico; (2) Dan McNeil, AFT Legal Department; (3) Ellen Bernstein, ATF President; (4) Shane Youtz, Attorney at Law; and (5) me 😉

A Labor Day Tribute to NY Teacher Sheri Lederman

Posted on by
Reply

As per the U.S. Department of Labor, “Labor Day, the first Monday in September, is a creation of the labor movement and is dedicated to the social and economic achievements of American workers. It constitutes a yearly national tribute to the contributions workers have made to the strength, prosperity, and well-being of our country.”

There is no better day than today, then, to honor Sheri Lederman – the Long Island, New York teacher who by all accounts other than her 1 out of 20 VAM score is a terrific 4th grade and now 18 year veteran teacher, and who along with her attorney husband, Bruce Lederman, is suing the state of New York to challenge the state’s teacher evaluation system…and mainly it’s value-added component (see prior posts about Sheri’s case here, here and here). I’m serving as one of the expert witnesses, also, on this case.

Anyhow, Al Jazeera America just released a fabulous 3:41 video about Sheri, her story, and VAMs in general. Columbia Professor Aaron Pallas is featured speaking on Sheri’s behalf, and a study of mine is featured at marker 2:41 (i.e., the national map illustrating what’s happening across the nation; see also “Putting growth and value-added models on the map: A national overview.”

Do give this a watch. It’s well worth honoring and respecting Sheri’s achievements, thus far, and especially today.

WATCH THE FULL VIDEO HERE!!

Stanford Professor Haertel: Short Video about VAM Reliability and Bias

Posted on by
1

I just came across this 3-minute video that you all might/should find of interest (click here for direct link to this video on YouTube; click here to view the video’s original posting on Stanford’s Center for Opportunity Policy in Education (SCOPE)).

Featured is Stanford’s Professor Emeritus – Dr. Edward Haertel – describing what he sees as two major flaws in the use of VAMs for teacher evaluation and accountability. These are two flaws serious enough, he argues, to prevent others from using VAM scores to make high-stakes decisions about really any of America’s public school teachers. “Like all measurements, these scores are imperfect. They are appropriate and useful for some purposes, but not for others. Viewed from a measurement perspective, value-added scores have limitations that make them unsuitable for high-stakes personnel decisions.”

The first problem is the unreliability of VAM scores which is attributed to noise from the data. The effect of a teacher is important, but weak when all of the other contributing factors are taken into account. The separation of the effect of a teacher from all the other effects is very difficult. This isn’t a flaw that can be fixed by more sophisticated statistical models; it is innate to the data collected.

The second problem is that the models must account for bias. The bias is the difference in circumstances faced by a teacher in a strong school and a teacher in a high-needs school. The instructional history of a student includes out of school support, peer support, and the academic learning climate of the school and VAMs do not take these important factors into account.