Learning from What Doesn’t Work in Teacher Evaluation

One of my doctoral students — Kevin Close — and I just had a study published in the practitioner journal Phi Delta Kappan that I wanted to share out with all of you, especially before the study is no longer open-access or free (see full study as currently available here). As the title indicates, the study is about how states, school districts, and schools can “Learn from What Doesn’t Work in Teacher Evaluation,” given an analysis that the two of us conducted of all documents pertaining to the four teacher evaluation and value-added model (VAM)-centered lawsuits in which I have been directly involved, and that I have also covered in this blog. These lawsuits include Lederman v. King in New York (see here), American Federation of Teachers et al. v. Public Education Department in New Mexico (see here), Houston Federation of Teachers v. Houston Independent School District in Texas (see here), and Trout v. Knox County Board of Education in Tennessee (see here).

Via this analysis we set out to comb through the legal documents to identify the strongest objections, as also recognized by the courts in these lawsuits, to VAMs as teacher measurement and accountability strategies. “The lessons to be learned from these cases are both important and timely” given that “[u]nder the Every Student Succeeds Act (ESSA), local education leaders once again have authority to decide for themselves how to assess teachers’ work.”

The most pertinent and also common issues as per these cases were as follows:

(1) Inconsistencies in teachers’ VAM-based estimates from one year to the next that are sometimes “wildly different.” Across these lawsuits, issues with reliability were very evident, whereas teachers classified as “effective” one year were either theorized or demonstrated to have around a 25%-59% chance of being classified as “ineffective” the next year, or vice versa, with other permutations also possible. As per our profession’s Standards for Educational and Psychological Testing, reliability should, rather, be observed whereby VAM estimates of teacher effectiveness are more or less consistent over time, from one year to the next, regardless of the type of students and perhaps subject areas that teachers teach.

(2) Bias in teachers’ VAM-based estimates were also of note, whereby documents suggested or evidenced that bias, or rather biased estimates of teachers’ actual effects does indeed exist (although this area was also of most contention and dispute). Specific to VAMs, since teachers are not randomly assigned the students they teach, whether their students are invariably more or less motivated, smart, knowledgeable, or capable can bias students’ test-based data, and teachers’ test-based data when aggregated. Court documents, although again not without counterarguments, suggested that VAM-based estimates are sometimes biased, especially when relatively homogeneous sets of students (i.e., English Language Learners (ELLs), gifted and special education students, free-or-reduced lunch eligible students) are non-randomly concentrated into schools, purposefully placed into classrooms, or both. Research suggests that this also sometimes happens regardless of the the sophistication of the statistical controls used to block said bias.

(3) The gaming mechanisms in play within teacher evaluation systems in which VAMs play a key role, or carry significant evaluative weight, were also of legal concern and dispute. That administrators sometimes inflate the observational ratings of their teachers whom they want to protect, while simultaneously offsetting the weight the VAMs sometimes carry was of note, as was the inverse. That administrators also sometimes lower teachers’ ratings to better align them with their “more objective” VAM counterparts were also at issue. “So argued the plaintiffs in the Houston and Tennessee lawsuits, for example. In those systems, school leaders appear to have given precedence to VAM scores, adjusting their classroom observations to match them. In both cases, administrators admitted to doing so, explaining that they sensed pressure to ensure that their ‘subjective’ classroom ratings were in sync with the VAM’s ‘objective’ scores.” Both sets of behavior distort the validity (or “truthfulness”) of any teacher evaluation system and are in violation of the same, aforementioned Standards for Educational and Psychological Testing that call for VAM scores and observation ratings to be kept separate. One indicator should never be adjusted to offset or to fit the other.

(4) Transparency, or the lack thereof, was also a common issue across cases. Transparency, which can be defined as the extent to which something is accessible and readily capable of being understood, pertains to whether VAM-based estimates are accessible and make sense to those at the receiving ends. “Not only should [teachers] have access to [their VAM-based] information for instructional purposes, but if they believe their evaluations to be unfair, they should be able to see all of the relevant data and calculations so that they can defend themselves.” In no case was this more legally pertinent than in Houston Federation of Teachers v. Houston Independent School District in Texas. Here, the presiding judge ruled that teachers did have “legitimate claims to see how their scores were calculated. Concealing this information, the judge ruled, violated teachers’ due process protections under the 14th Amendment (which holds that no state — or in this case organization — shall deprive any person of life, liberty, or property, without due process). Given this precedent, it seems likely that teachers in other states and districts will demand transparency as well.”

In the main article (here) we also discuss what states are now doing to (hopefully) improve upon their teacher evaluation systems in terms of using multiple measures to help to evaluate teachers more holistically. We emphasize the (in)formative versus the summative and high-stakes functions of such systems, and allowing teachers to take ownership over such systems in their development and implementation. I will leave you all to read the full article (here) for these details.

In sum, though, when rethinking states’ teacher evaluation systems, especially given the new liberties afforded to states via the Every Student Succeeds Act (ESSA), educators, education leaders, policymakers, and the like would do well to look to the past for guidance on what not to do — and what to do better. These legal cases can certainly inform such efforts.

Reference: Close, K., & Amrein-Beardsley, A. (2018). Learning from what doesn’t work in teacher evaluation. Phi Delta Kappan, 100(1), 15-19. Retrieved from http://www.kappanonline.org/learning-from-what-doesnt-work-in-teacher-evaluation/

Can More Teachers Be Covered Using VAMs?

Some researchers continue to explore the potential worth of value-added models (VAMs) for measuring teacher effectiveness. Not that I endorse the perpetual tweaking of this or twisting of that to explore how VAMs might be made “better” for such purposes, also given the abundance of decades research we now have evidencing the plethora of problems with using VAMs for such purposes, I do try to write about current events including current research published on this topic for this blog. Hence, I write here about a study researchers from Mathematica Policy Research released last month, about whether more teachers might be VAM-eligible (download the full study here).

One of the main issues with VAMs is that they can typically be used to measure the effects of only approximately 30% of all public school teachers. The other 70%, which sometimes includes entire campuses of teachers (e.g., early elementary and high school teachers) or teachers who do not teach the core subject areas assessed using large-scale standardized tests (e.g., mathematics and reading/language arts) cannot be evaluated or held accountable using VAM data. This is more generally termed an issue with fairness, defined by our profession’s Standards for Educational and Psychological Testing as the impartiality of “test score interpretations for intended use(s) for individuals from all [emphasis added] relevant subgroups” (p. 219). Issues of fairness arise when a test, or test-based inference or use impacts some more than others in unfair or prejudiced, yet often consequential ways.

Accordingly, in this study researchers explored whether VAMs can be used to evaluate teachers of subject areas that are only tested occasionally and in non-consecutive grade levels (e.g., science and social studies, for example, in grades 4 and 7 or 5 and 8) using teachers’ students’ other, consecutively administered subject area tests (i.e., mathematics and reading/language arts) can be used to help isolate teachers’ contributions to students’ achievement in said excluded subject areas. Indeed, it is true that “states and districts have little information about how value-added models [VAMs] perform in grades when tests in the same subject are not available from the previous year.” Yet, states (e.g., New Mexico) continue to do this without evidence that it works. This is also one point of contention in the ongoing lawsuit there. Hence, the purpose of this study was to explore (using state-level data from Oklahoma) how well doing this works, again, given the use of such proxy pretests “could allow states and districts to increase the number of teachers for whom value-added models [could] be used” (i.e., increase fairness).

However, researchers found that when doing just this (1) VAM estimates that do not account for a same-subject pretests may be less credible than estimates that use same-subject pretests from prior and adjacent grade levels (note that authors do not explicitly define what they mean by credible but infer the term to be synonymous with valid). In addition, (2) doing this may subsequently lead to relatively more biased VAM estimates, even more so than changing some other features of VAMs, and (3) doing this may make VAM estimates less precise, or reliable. Put more succinctly, using mathematics and reading/language arts as pretest scores to help measure (e.g., science and social studies) teachers’ value-added effects yields VAM estimates that are less credible (aka less valid), more biased, and less precise (aka less reliable).

The authors conclude that “some policy makers might interpret [these] findings as firm evidence against using value-added estimates that rely on proxy pretests [may be] too strong. The choice between different evaluation measures always involves trade-offs, and alternatives to value-added estimates [e.g., classroom observations and student learning objectives {SLOs)] also have important limitations.”

Their suggestion, rather, is for “[p]olicymakers [to] reduce the weight given to value-added estimates from models that rely on proxy pretests relative to the weight given to those of other teachers in subjects with pretests.” With all of this, I disagree. Using this or that statistical adjustment, or shrinkage approach, or adjusted weights, or…etc., is as I said before, at this point frivolous.

Reference: Walsh, E., Dotter, D., & Liu, A. Y. (2018). Can more teachers be covered? The accuracy, credibility, and precision of value-added estimates with proxy pre-tests. Washington DC: Mathematica Policy Research. Retrieved from https://www.mathematica-mpr.com/our-publications-and-findings/publications/can-more-teachers-be-covered-the-accuracy-credibility-and-precision-of-value-added-estimates

A Win in New Jersey: Tests to Now Account for 5% of Teachers’ Evaluations

Phil Murphy, the Governor of New Jersey, is keeping his campaign promise to parents, students, and educators, according to a news article just posted by the New Jersey Education Association (NJEA; see here). As per the New Jersey Commissioner of Education – Dr. Lamont Repollet, who was a classroom teacher himself — throughout New Jersey, Partnership for Assessment of Readiness for College and Careers (PARCC) test scores will now account for just 5% of a teacher’s evaluation, which is down from 30% as mandated for approxunatelt five years prior by both Murphy’s and Repollet’s predecessors.

Alas, the New Jersey Department of Education and the Murphy administration have “shown their respect for the research.” Because state law continues to require that standardized test scores play some role in teacher evaluation, a decrease to 5% is a victory, perhaps with a revocation of this law forthcoming.

“Today’s announcement is another step by Gov. Murphy toward keeping a campaign promise to rid New Jersey’s public schools of the scourge of high-stakes testing. While tens of thousands of families across the state have already refused to subject their children to PARCC, schools are still required to administer it and educators are still subject to its arbitrary effects on their evaluation. By dramatically lowering the stakes for the test, Murphy is making it possible for educators and students alike to focus more time and attention on real teaching and learning.” Indeed, “this is a victory of policy over politics, powered by parents and educators.”

Way to go New Jersey!

ACT Also Finds but Discourages States’ Decreased Use of VAMs Post-ESSA

Last June (2018), I released a blog post covering three key findings from a study that I along with two others conducted on states’ revised teacher evaluation systems post the passage of the federal government’s Every Student Succeeds Act (ESSA; see the full study here). In short, we evidenced that (1) the role of growth or value-added models (VAMs) for teacher evaluation purposes is declining across states, (2) many states are embracing the increased local control afforded them via ESSA and no longer have one-size-fits-all teacher evaluation systems, and (3) the rhetoric surrounding teacher evaluation has changed (e.g., language about holding teachers accountable is increasingly less evident than language about teachers’ professional development and support).

Last week, a similar study was released by the ACT standardized testing company. As per the title of this report (see the full report here), they too found that there is a “Shrinking Use of Growth” across states’ “Teacher Evaluation Legislation since ESSA.” They also found that for some states there was a “complete removal” of the state’s teacher evaluation system (e.g., Maine, Washington), a postponement of the state’s teacher evaluation systems until further notice (e.g., Connecticut, Indiana, Tennessee) or a complete prohibition of the use of students’ standardized tests in any teachers’ evaluations moving forward (e.g., Connecticut, Idaho). Otherwise, as we also found, states are increasingly “allowing districts, rather than the state, to determine their evaluation frameworks” themselves.

Unlike in our study, however, ACT (perhaps not surprisingly as a standardized testing company that also advertises its tests’ value-added capacities; see, for example, here) cautions states against “the complete elimination of student growth as part of teacher evaluation systems” in that they have “clearly” (without citations in support) proven “their value and potential value.” Hence, ACT defines this as “a step backward.” In our aforementioned study and blog post we reviewed some of the actual evidence (with citations in support) and would, accordingly, contest all-day-long that these systems have a “clear” “proven” “value.” Hence, we (also perhaps not surprisingly) called our similar findings as “steps in the right direction.”

Regardless, in in all fairness, they recommend that states not “respond to the challenges of using growth
measures in evaluation systems by eliminating their use” but “first consider less drastic measures such as:

  • postponing the use of student growth for employment decisions while refinements to the system can be made;
  • carrying out special studies to better understand the growth model; and/or
  • reviewing evaluation requirements for teachers who teach in untested grades and subjects so that the measures used more accurately reflect their performance.

“Pursuing such refinements, rather than reversing efforts to make teacher evaluation more meaningful and reflective of performance, is the best first step toward improving states’ evaluation systems.”

Citation: Croft, M., Guffy, G., & Vitale, D. (2018). The shrinking use of growth: Teacher evaluation legislation since ESSA. Iowa City, IA: ACT. Retrieved from https://www.act.org/content/dam/act/unsecured/documents/teacher-evaluation-legislation-since-essa.pdf

States’ Teacher Evaluation Systems Moving in the “Right” Direction

Last week, a technical report that one of my current and one of my former doctoral students helped me to research and write, was published by the University of Colorado Boulder’s National Education Policy Center (NEPC). While you can navigate to and read the press release here, as well as download and read the full report here, I thought I would summarize the report’s most interesting facts in this post, for the readers/followers of this blog who are likely more interested in the findings pertaining to states’ revised teacher evaluation systems, post the federal passage of the Every Student Succeeds Act (ESSA).

In short, we collected and analyzed for purposes of this study the 51 (i.e., 50 states plus Washington DC) revised teacher evaluation plans submitted to the federal government post ESSA (i.e. spring/summer of 2017) We found, again as specific only to states’ teacher evaluation systems, three key findings:

— First, the role of growth or value-added models (VAMs) for teacher evaluation purposes is declining. That is, the number of states using statewide growth models or VAMs has decreased from 42% to 30% since 2014. This is certainly a step in the “right,” defined as research-informed, direction. See also Figure 1 below (Close, Amrein-Beardsley, & Collins, 2018, p. 13).

— Second, because ESSA loosened federal control of teacher evaluation, many states no longer have a one-size-fits-all teacher evaluation system. This is allowing local districts to make more choices about models, implementation, execution, and the like, in the contexts of the schools and communities in which schools exist.

— Third, the rhetoric surrounding teacher evaluation has changed: language about holding teachers accountable for their value-added effects, or lack thereof, is much less evident in post-ESSA plans. Rather, new plans make note of providing data to teachers as a means of supporting professional development and improvement, essentially shifting the purpose of the evaluation system away from summative and toward formative use.

We also set forth recommendations for states in this report, as based on the evidence noted above (and presented in much more detail in the full report). The recommendations that also directly pertain to states’ (and districts’) teacher evaluation systems are that states/districts:

  1. Take advantage of decreased federal control by formulating revised assessment policies informed by the viewpoints of as many stakeholders as feasible. Such informed revision can help remedy earlier weaknesses, promote effective implementation, stress correct interpretation, and yield formative information.
  2. Ensure that teacher evaluation systems rely on a balanced system of multiple measures, without disproportionate weight assigned to any one measure as allegedly “superior” than any other. If measures contradict one another, however, output from all measures should be interpreted judiciously.
  3. Emphasize data useful as formative feedback in state systems, so that specific weaknesses in student learning can be identified, targeted and used to inform teachers’ professional development.
  4. Mandate ongoing research and evaluation of state assessment systems and ensure that adequate resources are provided to support [ongoing] evaluation [efforts].
  5. Set goals for reducing proficiency gaps and outline procedures for developing strategies to effectively reduce gaps once they have been identified.

We hope this information helps, especially the states and districts still looking to other states to see what is trending. While we note in the title of this blog post as well as the title of the full report that all of this represents “some steps in the right direction,” there is still much work to be done. This is especially true in states, for example like New Mexico (see my most recent post about the ongoing lawsuit in this state here) and other states which have yet to give up on the false promises and limited research of such educational policies established almost one decade ago (e.g., Race to the Top; Duncan, 2009).

Citations:

Close, K., Amrein-Beardsley, A., & Collins, C. (2018). State-level assessments and teacher evaluation systems after the passage of the Every Student Succeeds Act: Some steps in the right direction. Boulder, CO: Nation Education Policy Center (NEPC). Retrieved from http://nepc.colorado.edu/publication/state-assessment

Duncan, A. (2009, July 4). The race to the top begins: Remarks by Secretary Arne Duncan. Retrieved from http://www.ed.gov/news/speeches/2009/07/07242009.html

New Mexico Teacher Evaluation Lawsuit Updates

In December of 2015 in New Mexico, via a preliminary injunction set forth by state District Judge David K. Thomson, all consequences attached to teacher-level value-added model (VAM) scores (e.g., flagging the files of teachers with low VAM scores) were suspended throughout the state until the state (and/or others external to the state) could prove to the state court that the system was reliable, valid, fair, uniform, and the like. The trial during which this evidence is to be presented by the state is currently set for this October. See more information about this ruling here.

As the expert witness for the plaintiffs in this case, I was deposed a few weeks ago here in Phoenix, given my analyses of the state’s data (supported by one of my PhD students – Tray Geiger). In short, we found and I testified during the deposition that:

  • In terms of uniformity and fairness, there seem to be 70% or so of New Mexico teachers who are ineligible to be assessed using VAMs, and this proportion held constant across the years of data analyzed. This is even more important to note knowing that when VAM-based data are to be used to make consequential decisions about teachers, issues with fairness and uniformity become even more important given accountability-eligible teachers are also those who are relatively more likely to realize the negative or reap the positive consequences attached to VAM-based estimates.
  • In terms of reliability (or the consistency of teachers’ VAM-based scores over time), approximately 40% of teachers differed by one quintile (quintiles are derived when a sample or population is divided into fifths) and approximately 28% of teachers differed, from year-to-year, by two or more quintiles in terms of their VAM-derived effectiveness ratings. These results make sense when New Mexico’s results are situated within the current literature, whereas teachers classified as “effective” one year can have a 25%-59% chance of being classified as “ineffective” the next, or vice versa, with other permutations also possible.
  • In terms of validity (i.e., concurrent related evidence of validity), and importantly as also situated within the current literature, the correlations between New Mexico teachers’ VAM-based and observational scores ranged from r = 0.153 to r = 0.210. Not only are these correlations very weak[1], they are also very weak as appropriately situated within the literature, via which it is evidenced that correlations between multiple VAMs and observational scores typically range from 0.30 ≤ r ≤ 0.50.
  • In terms of bias, New Mexico’s Caucasian teachers had significantly higher observation scores than non-Caucasian teachers implying, also as per the current research, that Caucasian teachers may be (falsely) perceived as being better teachers than non-Caucasians teachers given bias within these instruments and/or bias of the scorers observing and scoring teachers using these instruments in practice. See prior posts about observational-based bias here, here and here.
  • Also of note in terms of bias was that: (1) teachers with fewer years of experience yielded VAM scores that were significantly lower than teachers with more years of experience, with similar patterns noted across teachers’ observation scores, which could all mean, as also in line with common sense as well as the research, that teachers with more experience are typically better teachers; (2) teachers who taught English language learners (ELLs) or special education students had lower VAM scores across the board than those who did not teach such students; (3) teachers who taught gifted students had significantly higher VAM scores than non-gifted teachers which runs counter to the current research evidencing that teachers’ gifted students oft-thwart or prevent them from demonstrating growth given ceiling effects; (4) teachers in schools with lower relative proportions of ELLs, special education students, students eligible for free-or-reduced lunches, and students from racial minority backgrounds, as well as higher relative proportions of gifted students, consistently had significantly higher VAM scores. These results suggest that teachers in these schools are as a group better, and/or that VAM-based estimates might be biased against teachers not teaching in these schools, preventing them from demonstrating comparable growth.

To read more about the data and methods used, as well as other findings, please see my affidavit submitted to the court attached here: Affidavit Feb2018.

Although, also in terms of a recent update, I should also note that a few weeks ago, as per an article in the AlbuquerqueJournal, New Mexico’s teacher evaluation systems is now likely to be overhauled, or simply “expired” as early as 2019. In short, “all three Democrats running for governor and the lone Republican candidate…have expressed misgivings about using students’ standardized test scores to evaluate the effectiveness of [New Mexico’s] teachers, a key component of the current system [at issue in this lawsuit and] imposed by the administration of outgoing Gov. Susana Martinez.” All four candidates described the current system “as fundamentally flawed and said they would move quickly to overhaul it.”

While I/we will proceed our efforts pertaining to this lawsuit until further notice, this is also important to note at this time in that it seems that New Mexico’s policymakers of new are going to be much wiser than those of late, at least in these regards.

[1] Interpreting r: 0.8 ≤ r ≤ 1.0 = a very strong correlation; 0.6 ≤ r ≤ 0.8 = a strong correlation; 0.4 ≤ r ≤ 0.6 = a moderate correlation; 0.2 ≤ r ≤ 0.4 = a weak correlation; and 0.0 ≤ r ≤ 0.2 = a very weak correlation, if any at all.

 

Identifying Effective Teacher Preparation Programs Using VAMs Does Not Work

A New Study [does not] Show Why It’s So Hard to Improve Teacher Preparation” Programs (TPPs). More specifically, it shows why using value-added models (VAMs) to evaluate TPPs, and then ideally improving them using the value-added data derived, is nearly if not entirely impossible.

This is precisely why yet another, perhaps, commonsensical but highly improbable federal policy move to imitate great teacher education programs and shut down ineffective ones, as based on their graduates’ students test-based performance over time (i.e., value-added) continues to fail.

Accordingly, in another, although not-yet peer-reviewed or published study referenced in the article above, titled “How Much Does Teacher Quality Vary Across Teacher Preparation Programs? Reanalyzing Estimates from [Six] States,” authors Paul T. von Hippel, from the University of Texas at Austin, and Laura Bellows, a PhD Student from Duke University, investigated “whether the teacher quality differences between TPPs are large enough to make [such] an accountability system worthwhile” (p. 2). More specifically, using a meta-analysis technique, they reanalyzed the results of such evaluations in six of the approximately 16 states doing this (i.e., in New York, Louisiana, Missouri, Washington, Texas, and Florida), each of which ultimately yielded a peer-reviewed publication, and they found “that teacher quality differences between most TPPs [were] negligible [at approximately] 0-0.04 standard deviations in student test scores” (p. 2).

They also highlight some of the statistical practices that exaggerated the “true” differences noted between TPPs in each of these but also these types of studies in general, and consequently conclude that the “results of TPP evaluations in different states may vary not for substantive reasons, but because of the[se] methodological choices” (p. 5). Likewise, as is the case with value-added research in general, when “[f]aced with the same set of results, some authors may [also] believe they see intriguing differences between TPPs, while others may believe there is not much going on” (p. 6). With that being said, I will not cover these statistical/technical issue more here. Do read the full study for these details, though, as also important.

Related, they found that in every state, the variation that they statistically observed was greater among relatively small TPPs versus large ones. They suggest that this occurs, accordingly, due to estimation or statistical methods that may be inadequate for the task at hand. However, if this is true this also means that because there is relatively less variation observed among large TPPs, it may be much more difficult “to single out a large TPP that is significantly better or worse than average” (p. 30). Accordingly, there are
several ways to mistakenly single out a TPP as exceptional or less than, merely given TPP size. This is obviously problematic.

Nonetheless, the authors also note that before they began this study, in Missouri, Texas, and Washington, that “the differences between TPPs appeared small or negligible” (p. 29), but in Louisiana and New York “they appeared more substantial” (p. 29). After their (re)analyses, however, their found that the results from and across these six different states were “more congruent” (p. 29), as also noted prior (i.e., differences between TPPs around 0 and 0.04 SDs in student test scores).

“In short,” they conclude, that “TPP evaluations may have some policy value, but the value is more modest than was originally envisioned. [Likewise, it] is probably not meaningful to rank all the TPPs in a state; the true differences between most TPPs are too small to matter, and the estimated differences consist mostly of noise” (p. 29). As per the article cited prior, they added that “It appears that differences between [programs] are rarely detectable, and that if they could be detected they would usually be too small to support effective policy decisions.”

To see a study similar to this, that colleagues and I conducted in Arizona, and that was recently published in Teaching Education, see “An Elusive Policy Imperative: Data and Methodological Challenges When Using Growth in Student Achievement to Evaluate Teacher Education Programs’ ‘Value-Added” summarized and referenced here.

Bias in VAMs, According to Validity Expert Michael T. Kane

During the still ongoing, value-added lawsuit in New Mexico (see my most recent update about this case here), I was honored to testify as the expert witness on behalf of the plaintiffs (see, for example, here). I was also fortunate to witness the testimony of the expert witness who testified on behalf of the defendants – Thomas Kane, Economics Professor at Harvard and former Director of the Bill & Melinda Gates Foundation’s Measures of Effective Teaching (MET) studies. During Kane’s testimony, one of the highlights (i.e., for the plaintiffs), or rather the low-lights (i.e., for him and the defendants), in my opinion, was when one of the plaintiff’s attorney’s questioned Kane, on the stand, about his expertise in the area of validity. In sum, Kane responded that he defined himself as an “expert” in the area, having also been trained by some of the best. Consequently, the plaintiff’s attorney’s questioned Kane about different types of validity evidences (e.g., construct, content, criterion), and Kane could not answer those questions. The only form of validity evidence with which he was familiar, and which he could clearly define, was evidence related to predictive validity. This hardly made him the expert he proclaimed himself to be minutes prior.

Let’s not mince words, though, or in this case names.

A real expert in validity (and validity theory) is another Kane, who goes by the full name of Michael T. Kane. This Kane is The Samuel J. Messick Chair in Test Validity at the Educational Testing Service (ETS); this Kane wrote one of the best, most contemporary, and currently most foundational papers on validity (see here); and this Kane just released an ETS-sponsored paper on Measurement Error and Bias in Value-Added Models certainly of interest here. I summarize this piece below (see the PDF of this report here).

In this paper Kane examines “the origins of [value-added model (VAM)-based] bias and its potential impact” and indicates that bias that is observed “is an increasing linear function of the student’s prior achievement and can be quite large (e.g., half a true-score standard deviation) for very low-scoring and high-scoring students [i.e., students in the extremes of any normal distribution]” (p. 1). Hence, Kane argues, “[t]o the extent that students with relatively low or high prior scores are clustered in particular classes and schools, the student-level bias will tend to generate bias in VAM estimates of teacher and school effects” (p. 1; see also prior posts about this type of bias here, here, and here; see also Haertel (2013) cited below). Kane concludes that “[a]djusting for this bias is possible, but it requires estimates of generalizability (or reliability) coefficients that are more accurate and precise than those that are generally available for standardized achievement tests” (p. 1; see also prior posts about issues with reliability across VAMs here, here, and here).

Kane’s more specific points of note:

  • To accurately calculate teachers’/schools’ value-added, “current and prior scores have to be on the same scale (or on vertically aligned scales) for the differences to make sense. Furthermore, the scale has to be an interval scale in the sense that a difference of a certain number of points has, at least approximately, the same meaning along the scale, so that it makes sense to compare gain scores from different parts of the scale…some uncertainty about scale characteristics is not a problem for many applications of vertical scaling, but it is a serious problem if the proposed use of the scores (e.g., educational accountability based on growth scores) demands that the vertical scale be demonstrably equal interval” (p. 1).
  • Likewise, while some approaches can be used to minimize the need for such scales (e.g., residual gain scores, covariate-adjustment models, and ordinary least squares (OLS) regression approaches which are of specific interest in this piece), “it is still necessary to assume [emphasis added] that a difference of a certain number of points has more or less the same meaning along the score scale for the current test scores” (p. 2).
  • Related, “such adjustments can [still] be biased to the extent that the predicted score does not include all factors that may have an impact on student performance. Bias can also result from errors of measurement in the prior scores included in the prediction equation…[and this can be]…substantial” (p. 2).
  • Accordingly, “gains for students with high true scores on the prior year’s test will be overestimated, and the gains for students with low true scores in the prior year will be underestimated. To the extent that students with relatively low and high true scores tend to be clustered in particular classes and schools, the student-level bias will generate bias in estimates of teacher and school effects” (p. 2).
  • Hence, if not corrected, this source of bias could have a substantial negative impact on estimated VAM scores for teachers and schools that serve students with low prior true scores and could have a substantial positive impact for teachers and schools that serve mainly high-performing students” (p. 2).
  • Put differently, random errors in students’ prior scores may “tend to add a positive bias to the residual gain scores for students with prior scores above the population mean, and they [may] tend to add a negative bias to the residual gain scores for students with prior scores below the mean. Th[is] bias is associated with the well-known phenomenon of regression to the mean” (p. 10).
  • Although, at least this latter claim — that students with relatively high true scores in the prior year could substantially and positively impact their teachers’/schools value-added estimates — does run somewhat contradictory to other claims as evidenced in the literature in terms of the extent to which ceiling effects substantially and negatively impact their teachers’/schools value-added estimates (see, for example, Point #7 as per the ongoing lawsuit in Houston here, and see also Florida teacher Luke Flint’s “Story” here).
  • In sum, and as should be a familiar conclusion to followers of this blog, “[g]iven that the results of VAMs may be used for high-stakes decisions about teachers and schools in the context of accountability programs,…any substantial source of bias would be a matter of great concern” (p. 2).

Citation: Kane, M. T. (2017). Measurement error and bias in value-added models. Princeton, NJ: Educational Testing Service (ETS) Research Report Series. doi:10.1002/ets2.12153 Retrieved from http://onlinelibrary.wiley.com/doi/10.1002/ets2.12153/full

See also 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).

The New York Times on “The Little Known Statistician” Who Passed

As many of you may recall, I wrote a post last March about the passing of William L. Sanders at age 74. Sanders developed the Education Value-Added Assessment System (EVAAS) — the value-added model (VAM) on which I have conducted most of my research (see, for example, here and here) and the VAM at the core of most of the teacher evaluation lawsuits in which I have been (or still am) engaged (see here, here, and here).

Over the weekend, though, The New York Times released a similar piece about Sanders’s passing, titled “The Little-Known Statistician Who Taught Us to Measure Teachers.” Because I had multiple colleagues and blog followers email me (or email me about) this article, I thought I would share it out with all of you, with some additional comments, of course, but also given the comments I already made in my prior post here.

First, I will start by saying that the title of this article is misleading in that what this “little-known” statistician contributed to the field of education was hardly “little” in terms of its size and impact. Rather, Sanders and his associates at SAS Institute Inc. greatly influenced our nation in terms of the last decade of our nation’s educational policies, as largely bent on high-stakes teacher accountability for educational reform. This occurred in large part due to Sanders’s (and others’) lobbying efforts when the federal government ultimately choose to incentivize and de facto require that all states hold their teachers accountable for their value-added, or lack thereof, while attaching high-stakes consequences (e.g., teacher termination) to teachers’ value-added estimates. This, of course, was to ensure educational reform. This occurred at the federal level, as we all likely know, primarily via Race to the Top and the No Child Left Behind Waivers essentially forced upon states when states had to adopt VAMs (or growth models) to also reform their teachers, and subsequently their schools, in order to continue to receive the federal funds upon which all states still rely.

It should be noted, though, that we as a nation have been relying upon similar high-stakes educational policies since the late 1970s (i.e., for now over 35 years); however, we have literally no research evidence that these high-stakes accountability policies have yielded any of their intended effects, as still perpetually conceptualized (see, for example, Nevada’s recent legislative ruling here) and as still advanced via large- and small-scale educational policies (e.g., we are still A Nation At Risk in terms of our global competitiveness). Yet, we continue to rely on the logic in support of such “carrot and stick” educational policies, even with this last decade’s teacher- versus student-level “spin.” We as a nation could really not be more ahistorical in terms of our educational policies in this regard.

Regardless, Sanders contributed to all of this at the federal level (that also trickled down to the state level) while also actively selling his VAM to state governments as well as local school districts (i.e., including the Houston Independent School District in which teacher plaintiffs just won a recent court ruling against the Sanders value-added system here), and Sanders did this using sets of (seriously) false marketing claims (e.g., purchasing and using the EVAAS will help “clear [a] path to achieving the US goal of leading the world in college completion by the year 2020”). To see two empirical articles about the claims made to sell Sanders’s EVAAS system, the research non-existent in support of each of the claims, and the realities of those at the receiving ends of this system (i.e., teachers) as per their experiences with each of the claims, see here and here.

Hence, to assert that what this “little known” statistician contributed to education was trivial or inconsequential is entirely false. Thankfully, with the passage of the Every Student Succeeds Act” (ESSA) the federal government came around, in at least some ways. While not yet acknowledging how holding teachers accountable for their students’ test scores, while ideal, simply does not work (see the “Top Ten” reasons why this does not work here), at least the federal government has given back to the states the authority to devise, hopefully, some more research-informed educational policies in these regards (I know….).

Nonetheless, may he rest in peace (see also here), perhaps also knowing that his forever stance of “[making] no apologies for the fact that his methods were too complex for most of the teachers whose jobs depended on them to understand,” just landed his EVAAS in serious jeopardy in court in Houston (see here) given this stance was just ruled as contributing to the violation of teachers’ Fourteenth Amendment rights (i.e., no state or in this case organization shall deprive any person of life, liberty, or property, without due process [emphasis added]).

New Texas Lawsuit: VAM-Based Estimates as Indicators of Teachers’ “Observable” Behaviors

Last week I spent a few days in Austin, one day during which I provided expert testimony for a new state-level lawsuit that has the potential to impact teachers throughout Texas. The lawsuit — Texas State Teachers Association (TSTA) v. Texas Education Agency (TEA), Mike Morath in his Official Capacity as Commissioner of Education for the State of Texas.

The key issue is that, as per the state’s Texas Education Code (Sec. § 21.351, see here) regarding teachers’ “Recommended Appraisal Process and Performance Criteria,” The Commissioner of Education must adopt “a recommended teacher appraisal process and criteria on which to appraise the performance of teachers. The criteria must be based on observable, job-related behavior, including: (1) teachers’ implementation of discipline management procedures; and (2) the performance of teachers’ students.” As for the latter, the State/TEA/Commissioner defined, as per its Texas Administrative Code (T.A.C., Chapter 15, Sub-Chapter AA, §150.1001, see here), that teacher-level value-added measures should be treated as one of the four measures of “(2) the performance of teachers’ students;” that is, one of the four measures recognized by the State/TEA/Commissioner as an “observable” indicator of a teacher’s “job-related” performance.

While currently no district throughout the State of Texas is required to use a value-added component to assess and evaluate its teachers, as noted, the value-added component is listed as one of four measures from which districts must choose at least one. All options listed in the category of “observable” indicators include: (A) student learning objectives (SLOs); (B) student portfolios; (C) pre- and post-test results on district-level assessments; and (D) value-added data based on student state assessment results.

Related, the state has not recommended or required that any district, if the value-added option is selected, to choose any particular value-added model (VAM) or calculation approach. Nor has it recommended or required that any district adopt any consequences as attached to these output; however, things like teacher contract renewal and sharing teachers’ prior appraisals with other districts in which teachers might be applying for new jobs is not discouraged. Again, though, the main issue here (and the key points to which I testified) was that the value-added component is listed as an “observable” and “job-related” teacher effectiveness indicator as per the state’s administrative code.

Accordingly, my (5 hour) testimony was primarily (albeit among many other things including the “job-related” part) about how teacher-level value-added data do not yield anything that is observable in terms of teachers’ effects. Likewise, officially referring to these data in this way is entirely false, in fact, in that:

  • “We” cannot directly observe a teacher “adding” (or detracting) value (e.g., with our own eyes, like supervisors can when they conduct observations of teachers in practice);
  • Using students’ test scores to measure student growth upwards (or downwards) and over time, as is very common practice using the (very often instructionally insensitive) state-level tests required by No Child Left Behind (NCLB), and doing this once per year in mathematics and reading/language arts (that includes prior and other current teachers’ effects, summer learning gains and decay, etc.), is not valid practice. That is, doing this has not been validated by the scholarly/testing community; and
  • Worse and less valid is to thereafter aggregate this student-level growth to the teacher level and then call whatever “growth” (or the lack thereof) is because of something the teacher (and really only the teacher did), as directly “observable.” These data are far from assessing a teacher’s causal or “observable” impacts on his/her students’ learning and achievement over time. See, for example, the prior statement released about value-added data use in this regard by the American Statistical Association (ASA) here. In this statement it is written that: “Research on VAMs has been fairly consistent that aspects of educational effectiveness that are measurable and within teacher control represent a small part of the total variation [emphasis added to note that this is variation explained which = correlational versus causal research] in student test scores or growth; most estimates in the literature attribute between 1% and 14% of the total variability [emphasis added] to teachers. This is not saying that teachers have little effect on students, but that variation among teachers [emphasis added] accounts for a small part of the variation [emphasis added] in [said test] scores. The majority of the variation in [said] test scores is [inversely, 86%-99% related] to factors outside of the teacher’s control such as student and family background, poverty, curriculum, and unmeasured influences.”

If any of you have anything to add to this, please do so in the comments section of this post. Otherwise, I will keep you posted on how this goes. My current understanding is that this one will be headed to court.