“You Are More Than Your EVAAS Score!”

Justin Parmenter is a seventh-grade language arts teacher in Charlotte, North Carolina. Via his blog — Notes from the Chalkboard — he writes “musings on public education.” You can subscribe to his blog at the bottom of any of his blog pages, one of which I copied and pasted below for all of you following this blog (now at 43K followers!!).

His recent post is titled “Take heart, NC teachers. You are more than your EVAAS score!” and serves as a solid reminder of what teachers’ value-added scores (and namely in this case teachers’ Education Value-Added Assessment Scores (EVAAS)) cannot tell you, us, or pretty much anybody about anyone’s worth as a teacher. Do give it a read, and do give him a shout out by sharing this with others.

*****

Last night an email from the SAS corporation hit the inboxes of teachers all across North Carolina.  I found it suspicious and forwarded it to spam.

EVAAS is a tool that SAS claims shows how effective educators are by measuring precisely what value each teacher adds to their students’ learning.  Each year teachers board an emotional roller coaster as they prepare to find out whether they are great teachers, average teachers, or terrible teachers–provided they can figure out their logins.

NC taxpayers spend millions of dollars for this tool, and SAS founder and CEO James Goodnight is the richest person in North Carolina, worth nearly $10 billion.  However, over the past few years, more and more research has shown that value added ratings like EVAAS are highly unstable and are unable to account for the many factors that influence our students and their progress. Lawsuits have sprung up from Texas to Tennessee, charging, among other things, that use of this data to evaluate teachers and make staffing decisions violates teachers’ due process rights, since SAS refuses to reveal the algorithms it uses to calculate scores.

By coincidence, the same day I got the email from SAS, I also got this email from the mother of one of my 7th grade students:

Photos attached provided evidence that the student was indeed reading at the dinner table.

The student in question had never thought of himself as a reader.  That has changed this year–not because of any masterful teaching on my part, but just because he had the right book in front of him at the right time.

Here’s my point:  We need to remember that EVAAS can’t measure the most important ways teachers are adding value to our students’ lives.  Every day we are turning students into lifelong independent readers. We are counseling them through everything from skinned knees to school shootings.  We are mediating their conflicts. We are coaching them in sports. We are finding creative ways to inspire and motivate them. We are teaching them kindness and empathy.  We are doing so much more than helping them pass a standardized test at the end of the year.

So if you figure out your EVAAS login today, NC teachers, take heart.  You are so much more than your EVAAS score!

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

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).

A North Carolina Teacher’s Guest Post on His/Her EVAAS Scores

A teacher from the state of North Carolina recently emailed me for my advice regarding how to help him/her read and understand his/her recently received Education Value-Added Assessment System (EVAAS) value added scores. You likely recall that the EVAAS is the model I cover most on this blog, also in that this is the system I have researched the most, as well as the proprietary system adopted by multiple states (e.g., Ohio, North Carolina, and South Carolina) and districts across the country for which taxpayers continue to pay big $. Of late, this is also the value-added model (VAM) of sole interest in the recent lawsuit that teachers won in Houston (see here).

You might also recall that the EVAAS is the system developed by the now late William Sanders (see here), who ultimately sold it to SAS Institute Inc. that now holds all rights to the VAM (see also prior posts about the EVAAS here, here, here, here, here, and here). It is also important to note, because this teacher teaches in North Carolina where SAS Institute Inc. is located and where its CEO James Goodnight is considered the richest man in the state, that as a major Grand Old Party (GOP) donor “he” helps to set all of of the state’s education policy as the state is also dominated by Republicans. All of this also means that it is unlikely EVAAS will go anywhere unless there is honest and open dialogue about the shortcomings of the data.

Hence, the attempt here is to begin at least some honest and open dialogue herein. Accordingly, here is what this teacher wrote in response to my request that (s)he write a guest post:

***

SAS Institute Inc. claims that the EVAAS enables teachers to “modify curriculum, student support and instructional strategies to address the needs of all students.”  My goal this year is to see whether these claims are actually possible or true. I’d like to dig deep into the data made available to me — for which my state pays over $3.6 million per year — in an effort to see what these data say about my instruction, accordingly.

For starters, here is what my EVAAS-based growth looks like over the past three years:

As you can see, three years ago I met my expected growth, but my growth measure was slightly below zero. The year after that I knocked it out of the park. This past year I was right in the middle of my prior two years of results. Notice the volatility [aka an issue with VAM-based reliability, or consistency, or a lack thereof; see, for example, here].

Notwithstanding, SAS Institute Inc. makes the following recommendations in terms of how I should approach my data:

Reflecting on Your Teaching Practice: Learn to use your Teacher reports to reflect on the effectiveness of your instructional delivery.

The Teacher Value Added report displays value-added data across multiple years for the same subject and grade or course. As you review the report, you’ll want to ask these questions:

  • Looking at the Growth Index for the most recent year, were you effective at helping students to meet or exceed the Growth Standard?
  • If you have multiple years of data, are the Growth Index values consistent across years? Is there a positive or negative trend?
  • If there is a trend, what factors might have contributed to that trend?
  • Based on this information, what strategies and instructional practices will you replicate in the current school year? What strategies and instructional practices will you change or refine to increase your success in helping students make academic growth?

Yet my growth index values are not consistent across years, as also noted above. Rather, my “trends” are baffling to me.  When I compare those three instructional years in my mind, nothing stands out to me in terms of differences in instructional strategies that would explain the fluctuations in growth measures, either.

So let’s take a closer look at my data for last year (i.e., 2016-2017).  I teach 7th grade English/language arts (ELA), so my numbers are based on my students reading grade 7 scores in the table below.

What jumps out for me here is the contradiction in “my” data for achievement Levels 3 and 4 (achievement levels start at Level 1 and top out at Level 5, whereas levels 3 and 4 are considered proficient/middle of the road).  There is moderate evidence that my grade 7 students who scored a Level 4 on the state reading test exceeded the Growth Standard.  But there is also moderate evidence that my same grade 7 students who scored Level 3 did not meet the Growth Standard.  At the same time, the number of students I had demonstrating proficiency on the same reading test (by scoring at least a 3) increased from 71% in 2015-2016 (when I exceeded expected growth) to 76% in school year 2016-2017 (when my growth declined significantly). This makes no sense, right?

Hence, and after considering my data above, the question I’m left with is actually really important:  Are the instructional strategies I’m using for my students whose achievement levels are in the middle working, or are they not?

I’d love to hear from other teachers on their interpretations of these data.  A tool that costs taxpayers this much money and impacts teacher evaluations in so many states should live up to its claims of being useful for informing our teaching.

The More Weight VAMs Carry, the More Teacher Effects (Will Appear to) Vary

Matthew A. Kraft — an Assistant Professor of Education & Economics at Brown University and co-author of an article published in Educational Researcher on “Revisiting The Widget Effect” (here), and another of his co-authors Matthew P. Steinberg — an Assistant Professor of Education Policy at the University of Pennsylvania — just published another article in this same journal on “The Sensitivity of Teacher Performance Ratings to the Design of Teacher Evaluation Systems” (see the full and freely accessible, at least for now, article here; see also its original and what should be enduring version here).

In this article, Steinberg and Kraft (2017) examine teacher performance measure weights while conducting multiple simulations of data taken from the Bill & Melinda Gates Measures of Effective Teaching (MET) studies. They conclude that “performance measure weights and ratings” surrounding teachers’ value-added, observational measures, and student survey indicators play “critical roles” when “determining teachers’ summative evaluation ratings and the distribution of teacher proficiency rates.” In other words, the weighting of teacher evaluation systems’ multiple measures matter, matter differently for different types of teachers within and across school districts and states, and matter also in that so often these weights are arbitrarily and politically defined and set.

Indeed, because “state and local policymakers have almost no empirically based evidence [emphasis added, although I would write “no empirically based evidence”] to inform their decision process about how to combine scores across multiple performance measures…decisions about [such] weights…are often made through a somewhat arbitrary and iterative process, one that is shaped by political considerations in place of empirical evidence” (Steinberg & Kraft, 2017, p. 379).

This is very important to note in that the consequences attached to these measures, also given the arbitrary and political constructions they represent, can be both professionally and personally, career and life changing, respectively. How and to what extent “the proportion of teachers deemed professionally proficient changes under different weighting and ratings thresholds schemes” (p. 379), then, clearly matters.

While Steinberg and Kraft (2017) have other key findings they also present throughout this piece, their most important finding, in my opinion, is that, again, “teacher proficiency rates change substantially as the weights assigned to teacher performance measures change” (p. 387). Moreover, the more weight assigned to measures with higher relative means (e.g., observational or student survey measures), the greater the rate by which teachers are rated effective or proficient, and vice versa (i.e., the more weight assigned to teachers’ value-added, the higher the rate by which teachers will be rated ineffective or inadequate; as also discussed on p. 388).

Put differently, “teacher proficiency rates are lowest across all [district and state] systems when norm-referenced teacher performance measures, such as VAMs [i.e., with scores that are normalized in line with bell curves, with a mean or average centered around the middle of the normal distributions], are given greater relative weight” (p. 389).

This becomes problematic when states or districts then use these weighted systems (again, weighted in arbitrary and political ways) to illustrate, often to the public, that their new-and-improved teacher evaluation systems, as inspired by the MET studies mentioned prior, are now “better” at differentiating between “good and bad” teachers. Thereafter, some states over others are then celebrated (e.g., by the National Center of Teacher Quality; see, for example, here) for taking the evaluation of teacher effects more seriously than others when, as evidenced herein, this is (unfortunately) more due to manipulation than true changes in these systems. Accordingly, the fact remains that the more weight VAMs carry, the more teacher effects (will appear to) vary. It’s not necessarily that they vary in reality, but the manipulation of the weights on the back end, rather, cause such variation and then lead to, quite literally, such delusions of grandeur in these regards (see also here).

At a more pragmatic level, this also suggests that the teacher evaluation ratings for the roughly 70% of teachers who are not VAM eligible “are likely to differ in systematic ways from the ratings of teachers for whom VAM scores can be calculated” (p. 392). This is precisely why evidence in New Mexico suggests VAM-eligible teachers are up to five times more likely to be ranked as “ineffective” or “minimally effective” than their non-VAM-eligible colleagues; that is, “[also b]ecause greater weight is consistently assigned to observation scores for teachers in nontested grades and subjects” (p. 392). This also causes a related but also important issue with fairness, whereas equally effective teachers, just by being VAM eligible, may be five-or-so times likely (e.g., in states like New Mexico) of being rated as ineffective by the mere fact that they are VAM eligible and their states, quite literally, “value” value-added “too much” (as also arbitrarily defined).

Finally, it should also be noted as an important caveat here, that the findings advanced by Steinberg and Kraft (2017) “are not intended to provide specific recommendations about what weights and ratings to select—such decisions are fundamentally subject to local district priorities and preferences. (p. 379). These findings do, however, “offer important insights about how these decisions will affect the distribution of teacher performance ratings as policymakers and administrators continue to refine and possibly remake teacher evaluation systems” (p. 379).

Related, please recall that via the MET studies one of the researchers’ goals was to determine which weights per multiple measure were empirically defensible. MET researchers failed to do so and then defaulted to recommending an equal distribution of weights without empirical justification (see also Rothstein & Mathis, 2013). This also means that anyone at any state or district level who might say that this weight here or that weight there is empirically defensible should be asked for the evidence in support.

Citations:

Rothstein, J., & Mathis, W. J. (2013, January). Review of two culminating reports from the MET Project. Boulder, CO: National Educational Policy Center. Retrieved from http://nepc.colorado.edu/thinktank/review-MET-final-2013

Steinberg, M. P., & Kraft, M. A. (2017). The sensitivity of teacher performance ratings to the design of teacher evaluation systems. Educational Researcher, 46(7), 378–
396. doi:10.3102/0013189X17726752 Retrieved from http://journals.sagepub.com/doi/abs/10.3102/0013189X17726752

Breaking News: The End of Value-Added Measures for Teacher Termination in Houston

Recall from multiple prior posts (see, for example, here, here, here, here, and here) that a set of teachers in the Houston Independent School District (HISD), with the support of the Houston Federation of Teachers (HFT) and the American Federation of Teachers (AFT), took their district to federal court to fight against the (mis)use of their value-added scores derived via the Education Value-Added Assessment System (EVAAS) — the “original” value-added model (VAM) developed in Tennessee by William L. Sanders who just recently passed away (see here). Teachers’ EVAAS scores, in short, were being used to evaluate teachers in Houston in more consequential ways than any other district or state in the nation (e.g., the termination of 221 teachers in one year as based, primarily, on their EVAAS scores).

The case — Houston Federation of Teachers et al. v. Houston ISD — was filed in 2014 and just one day ago (October 10, 2017) came the case’s final federal suit settlement. Click here to read the “Settlement and Full and Final Release Agreement.” But in short, this means the “End of Value-Added Measures for Teacher Termination in Houston” (see also here).

More specifically, recall that the judge notably ruled prior (in May of 2017) that the plaintiffs did have sufficient evidence to proceed to trial on their claims that the use of EVAAS in Houston to terminate their contracts was a violation of their Fourteenth Amendment due process protections (i.e., no state or in this case district shall deprive any person of life, liberty, or property, without due process). That is, the judge ruled that “any effort by teachers to replicate their own scores, with the limited information available to them, [would] necessarily fail” (see here p. 13). This was confirmed by the one of the plaintiffs’ expert witness who was also “unable to replicate the scores despite being given far greater access to the underlying computer codes than [was] available to an individual teacher” (see here p. 13).

Hence, and “[a]ccording to the unrebutted testimony of [the] plaintiffs’ expert [witness], without access to SAS’s proprietary information – the value-added equations, computer source codes, decision rules, and assumptions – EVAAS scores will remain a mysterious ‘black box,’ impervious to challenge” (see here p. 17). Consequently, the judge concluded that HISD teachers “have no meaningful way to ensure correct calculation of their EVAAS scores, and as a result are unfairly subject to mistaken deprivation of constitutionally protected property interests in their jobs” (see here p. 18).

Thereafter, and as per this settlement, HISD agreed to refrain from using VAMs, including the EVAAS, to terminate teachers’ contracts as long as the VAM score is “unverifiable.” More specifically, “HISD agree[d] it will not in the future use value-added scores, including but not limited to EVAAS scores, as a basis to terminate the employment of a term or probationary contract teacher during the term of that teacher’s contract, or to terminate a continuing contract teacher at any time, so long as the value-added score assigned to the teacher remains unverifiable. (see here p. 2; see also here). HISD also agreed to create an “instructional consultation subcommittee” to more inclusively and democratically inform HISD’s teacher appraisal systems and processes, and HISD agreed to pay the Texas AFT $237,000 in its attorney and other legal fees and expenses (State of Texas, 2017, p. 2; see also AFT, 2017).

This is yet another big win for teachers in Houston, and potentially elsewhere, as this ruling is an unprecedented development in VAM litigation. Teachers and others using the EVAAS or another VAM for that matter (e.g., that is also “unverifiable”) do take note, at minimum.

On Conditional Bias and Correlation: A Guest Post

After I posted about “Observational Systems: Correlations with Value-Added and Bias,” a blog follower, associate professor, and statistician named Laura Ring Kapitula (see also a very influential article she wrote on VAMs here) posted comments on this site that I found of interest, and I thought would also be of interest to blog followers. Hence, I invited her to write a guest post, and she did.

She used R (i.e., a free software environment for statistical computing and graphics) to simulate correlation scatterplots (see Figures below) to illustrate three unique situations: (1) a simulation where there are two indicators (e.g., teacher value-added and observational estimates plotted on the x and y axes) that have a correlation of r = 0.28 (the highest correlation coefficient at issue in the aforementioned post); (2) a simulation exploring the impact of negative bias and a moderate correlation on a group of teachers; and (3) another simulation with two indicators that have a non-linear relationship possibly induced or caused by bias. She designed simulations (2) and (3) to illustrate the plausibility of the situation suggested next (as written into Audrey’s post prior) about potential bias in both value-added and observational estimates:

If there is some bias present in value-added estimates, and some bias present in the observational estimates…perhaps this is why these low correlations are observed. That is, only those teachers teaching classrooms inordinately stacked with students from racial minority, poor, low achieving, etc. groups might yield relatively stronger correlations between their value-added and observational scores given bias, hence, the low correlations observed may be due to bias and bias alone.

Laura continues…

Here, Audrey makes the point that a correlation of r = 0.28 is “weak.” It is, accordingly, useful to see an example of just how “weak” such a correlation is by looking at a scatterplot of data selected from a population where the true correlation is r = 0.28. To make the illustration more meaningful the points are colored based on their quintile scores as per simulated teachers’ value-added divided into the lowest 20%, next 20%, etc.

In this figure you can see by looking at the blue “least squares line” that, “on average,” as a simulated teacher’s value-added estimate increases the average of a teacher’s observational estimate increases. However, there is a lot of variability (or scatter points) around the (scatterplot) line. Given this variability, we can make statements about averages, such as “on average” teachers in the top 20% for VAM scores will likely have on average higher observed observational scores; however, there is not nearly enough precision to make any (and certainly not any good) predictions about the observational score from the VAM score for individual teachers. In fact, the linear relationship between teachers’ VAM and observational scores only accounts for about 8% of the variation in VAM score. Note: we get 8% by squaring the aforementioned r = 0.28 correlation (i.e., an R squared). The other 92% of the variance is due to error and other factors.

What this means in practice is that when correlations are this “weak,” it is reasonable to say statements about averages, for example, that “on average” as one variable increases the mean of the other variable increases, but it would not be prudent or wise to make predictions for individuals based on these data. See, for example, that individuals in the top 20% (quintile 5) of VAM have a very large spread in their scores on the observational score, with 95% of the scores in the top quintile being in between the 7th and 98th percentiles for their observational scores. So, here if we observe a VAM for a specific teacher in the top 20%, and we do not know their observational score, we cannot say much more than their observational score is likely to be in the top 90%. Similarly, if we observe a VAM in the bottom 20%, we cannot say much more than their observational score is likely to be somewhere in the bottom 90%. That’s not saying a lot, in terms of precision, but also in terms of practice.

The second scatterplot I ran to test how bias that only impacts a small group of teachers might theoretically impact an overall correlation, as posited by Audrey. Here I simulated a situation where, again, there are two values present in a population of teachers: a teacher’s value-added and a teacher’s observational score. Then I insert a group of teachers (as Audrey described) who represent 20% of a population and teach a disproportionate number of students who come from relatively lower socioeconomic, high racial minority, etc. backgrounds, and I assume this group is measured with negative bias on both indicators and this group has a moderate correlation between indicators of r = 0.50. The other 80% of the population is assumed to be uncorrelated. Note: for this demonstration I assume that this group includes 20% of teachers from the aforementioned population, these teachers I assume to be measured with negative bias (by one standard deviation on average) on both measures, and, again, I set their correlation at r = 0.50 with the other 80% of teachers at a correlation of zero.

What you can see is that if there is bias in this correlation that impacts only a certain group on the two instrument indicators; hence, it is possible that this bias can result in an observed correlation overall. In other words, a strong correlation noted in just one group of teachers (i.e., teachers scoring the lowest on their value-added and observational indicators in this case) can be relatively stronger than the “weak” correlation observed on average or overall.

Another, possible situation is that there might be a non-linear relationship between these two measures. In the simulation below, I assume that different quantiles on VAM have a different linear relationship with the observational score. For example, in the plot there is not a constant slope, but teachers who are in the first quintile on VAM I assume to have a correlation of r = 0.50 with observational scores, the second quintile I assume to have a correlation of r = 0.20, and the other quintiles I assume to be uncorrelated. This results in an overall correlation in the simulation of r = 0.24, with a very small p-value (i.e. a very small chance that a correlation of this size would be observed by random chance alone if the true correlation was zero).

What this means in practice is that if, in fact, there is a non-linear relationship between teachers’ observational and VAM scores, this can induce a small but statistically significant correlation. As evidenced, teachers in the lowest 20% on the VAM score have differences in the mean observational score depending on the VAM score (a moderate correlation of r = 0.50), but for the other 80%, knowing the VAM score is not informative as there is a very small correlation for the second quintile and no correlation for the upper 60%. So, if quintile cut-off scores are used, teachers can easily be misclassified. In sum, Pearson Correlations (the standard correlation coefficient) measure the overall strength of  linear relationships between X and Y, but if X and Y have a non-linear relationship (like as illustrated in the above), this statistic can be very misleading.

Note also that for all of these simulations very small p-values are observed (e.g., p-values <0.0000001 which, again, mean these correlations are statistically significant or that the probability of observing correlations this large by chance if the true correlation is zero, is nearly 0%). What this illustrates, again, is that correlations (especially correlations this small) are (still) often misleading. While they might be statistically significant, they might mean relatively little in the grand scheme of things (i.e., in terms of practical significance; see also “The Difference Between”Significant’ and ‘Not Significant’ is not Itself Statistically Significant” or posts on Andrew Gelman’s blog for more discussion on these topics if interested).

At the end of the day r = 0.28 is still a “weak” correlation. In addition, it might be “weak,” on average, but much stronger and statistically and practically significant for teachers in the bottom quintiles (e.g., teachers in the bottom 20%, as illustrated in the final figure above) typically teaching the highest needs students. Accordingly, this might be due, at least in part, to bias.

In conclusion, one should always be wary of claims based on “weak” correlations, especially if they are positioned to be stronger than industry standards would classify them (e.g., in the case highlighted in the prior post). Even if a correlation is “statistically significant,” it is possible that the correlation is the result of bias, and that the relationship is so weak that it is not meaningful in practice, especially when the goal is to make high-stakes decisions about individual teachers. Accordingly, when you see correlations this small, keep these scatterplots in mind or generate some of your own (see, for example, here to dive deeper into what these correlations might mean and how significant these correlations might really be).

*Please contact Dr. Kapitula directly at kapitull@gvsu.edu if you want more information or to access the R code she used for the above.

The “Widget Effect” Report Revisited

You might recall that in 2009, The New Teacher Project published a highly influential “Widget Effect” report in which researchers (see citation below) evidenced that 99% of teachers (whose teacher evaluation reports they examined across a sample of school districts spread across a handful of states) received evaluation ratings of “satisfactory” or higher. Inversely, only 1% of the teachers whose reports researchers examined received ratings of “unsatisfactory,” even though teachers’ supervisors could identify more teachers whom they deemed ineffective when asked otherwise.

Accordingly, this report was widely publicized given the assumed improbability that only 1% of America’s public school teachers were, in fact, ineffectual, and given the fact that such ineffective teachers apparently existed but were not being identified using standard teacher evaluation/observational systems in use at the time.

Hence, this report was used as evidence that America’s teacher evaluation systems were unacceptable and in need of reform, primarily given the subjectivities and flaws apparent and arguably inherent across the observational components of these systems. This reform was also needed to help reform America’s public schools, writ large, so the logic went and (often) continues to go. While binary constructions of complex data such as these are often used to ground simplistic ideas and push definitive policies, ideas, and agendas, this tactic certainly worked here, as this report (among a few others) was used to inform the federal and state policies pushing teacher evaluation system reform as a result (e.g., Race to the Top (RTTT)).

Likewise, this report continues to be used whenever a state’s or district’s new-and-improved teacher evaluation systems (still) evidence “too many” (as typically arbitrarily defined) teachers as effective or higher (see, for example, an Education Week article about this here). Although, whether in fact the systems have actually been reformed is also of debate in that states are still using many of the same observational systems they were using prior (i.e., not the “binary checklists” exaggerated in the original as well as this report, albeit true in the case of the district of focus in this study). The real “reforms,” here, pertained to the extent to which value-added model (VAM) or other growth output were combined with these observational measures, and the extent to which districts adopted state-level observational models as per the centralized educational policies put into place at the same time.

Nonetheless, now eight years later, Matthew A. Kraft – an Assistant Professor of Education & Economics at Brown University and Allison F. Gilmour – an Assistant Professor at Temple University (and former doctoral student at Vanderbilt University), revisited the original report. Just published in the esteemed, peer-reviewed journal Educational Researcher (see an earlier version of the published study here), Kraft and Gilmour compiled “teacher performance ratings across 24 [of the 38, including 14 RTTT] states that [by 2014-2015] adopted major reforms to their teacher evaluation systems” as a result of such policy initiatives. They found that “the percentage of teachers rated Unsatisfactory remains less than 1%,” except for in two states (i.e., Maryland and New Mexico), with Unsatisfactory (or similar) ratings varying “widely across states with 0.7% to 28.7%” as the low and high, respectively (see also the study Abstract).

Related, Kraft and Gilmour found that “some new teacher evaluation systems do differentiate among teachers, but most only do so at the top of the ratings spectrum” (p. 10). More specifically, observers in states in which teacher evaluation ratings include five versus four rating categories differentiate teachers more, but still do so along the top three ratings, which still does not solve the negative skew at issue (i.e., “too many” teachers still scoring “too well”). They also found that when these observational systems were used for formative (i.e., informative, improvement) purposes, teachers’ ratings were lower than when they were used for summative (i.e., final summary) purposes.

Clearly, the assumptions of all involved in this area of policy research come into play, here, akin to how they did in The Bell Curve and The Bell Curve Debate. During this (still ongoing) debate, many fervently debated whether socioeconomic and educational outcomes (e.g., IQ) should be normally distributed. What this means in this case, for example, is that for every teacher who is rated highly effective there should be a teacher rated as highly ineffective, more or less, to yield a symmetrical distribution of teacher observational scores across the spectrum.

In fact, one observational system of which I am aware (i.e., the TAP System for Teacher and Student Advancement) is marketing its proprietary system, using as a primary selling point figures illustrating (with text explaining) how clients who use their system will improve their prior “Widget Effect” results (i.e., yielding such normal curves; see Figure below, as per Jerald & Van Hook, 2011, p. 1).

Evidence also suggests that these scores are also (sometimes) being artificially deflated to assist in these attempts (see, for example, a recent publication of mine released a few days ago here in the (also) esteemed, peer-reviewed Teachers College Record about how this is also occurring in response to the “Widget Effect” report and the educational policies that follows).

While Kraft and Gilmour assert that “systems that place greater weight on normative measures such as value-added scores rather than…[just]…observations have fewer teachers rated proficient” (p. 19; see also Steinberg & Kraft, forthcoming; a related article about how this has occurred in New Mexico here; and New Mexico’s 2014-2016 data below and here, as also illustrative of the desired normal curve distributions discussed above), I highly doubt this purely reflects New Mexico’s “commitment to putting students first.”

I also highly doubt that, as per New Mexico’s acting Secretary of Education, this was “not [emphasis added] designed with quote unquote end results in mind.” That is, “the New Mexico Public Education Department did not set out to place any specific number or percentage of teachers into a given category.” If true, it’s pretty miraculous how this simply worked out as illustrated… This is also at issue in the lawsuit in which I am involved in New Mexico, in which the American Federation of Teachers won an injunction in 2015 that still stands today (see more information about this lawsuit here). Indeed, as per Kraft, all of this “might [and possibly should] undercut the potential for this differentiation [if ultimately proven artificial, for example, as based on statistical or other pragmatic deflation tactics] to be seen as accurate and valid” (as quoted here).

Notwithstanding, Kraft and Gilmour, also as part (and actually the primary part) of this study, “present original survey data from an urban district illustrating that evaluators perceive more than three times as many teachers in their schools to be below Proficient than they rate as such.” Accordingly, even though their data for this part of this study come from one district, their findings are similar to others evidenced in the “Widget Effect” report; hence, there are still likely educational measurement (and validity) issues on both ends (i.e., with using such observational rubrics as part of America’s reformed teacher evaluation systems and using survey methods to put into check these systems, overall). In other words, just because the survey data did not match the observational data does not mean either is wrong, or right, but there are still likely educational measurement issues.

Also of issue in this regard, in terms of the 1% issue, is (a) the time and effort it takes supervisors to assist/desist after rating teachers low is sometimes not worth assigning low ratings; (b) how supervisors often give higher ratings to those with perceived potential, also in support of their future growth, even if current evidence suggests a lower rating is warranted; (c) how having “difficult conversations” can sometimes prevent supervisors from assigning the scores they believe teachers may deserve, especially if things like job security are on the line; (d) supervisors’ challenges with removing teachers, including “long, laborious, legal, draining process[es];” and (e) supervisors’ challenges with replacing teachers, if terminated, given current teacher shortages and the time and effort, again, it often takes to hire (ideally more qualified) replacements.

References:

Jerald, C. D., & Van Hook, K. (2011). More than measurement: The TAP system’s lessons learned for designing better teacher evaluation systems. Santa Monica, CA: National Institute for Excellence in Teaching (NIET). Retrieved from http://files.eric.ed.gov/fulltext/ED533382.pdf

Kraft, M. A, & Gilmour, A. F. (2017). Revisiting the Widget Effect: Teacher evaluation reforms and the distribution of teacher effectiveness. Educational Researcher, 46(5) 234-249. doi:10.3102/0013189X17718797

Steinberg, M. P., & Kraft, M. A. (forthcoming). The sensitivity of teacher performance ratings to the design of teacher evaluation systems. Educational Researcher.

Weisberg, D., Sexton, S., Mulhern, J., & Keeling, D. (2009). “The Widget Effect.” Education Digest, 75(2), 31–35.

The Tripod Student Survey Instrument: Its Factor Structure and Value-Added Correlations

The Tripod student perception survey instrument is a “research-based” instrument increasingly being used by states to add to state’s teacher evaluation systems as based on “multiple measures.” While there are other instruments also in use, as well as student survey instruments being developed by states and local districts, this one in particular is gaining in popularity, also in that it was used throughout the Bill & Melinda Gates Foundation’s ($43 million worth of) Measures of Effective Teaching (MET) studies. A current estimate (as per the study discussed in this post) is that during the 2015–2016 school year approximately 1,400 schools purchased and administered the Tripod. See also a prior post (here) about this instrument, or more specifically a chapter of a book about the instrument as authored by the instrument’s developer and lead researcher in a  research surrounding it – Ronald Ferguson.

In a study recently released in the esteemed American Educational Research Journal (AERJ), and titled “What Can Student Perception Surveys Tell Us About Teaching? Empirically Testing the Underlying Structure of the Tripod Student Perception Survey,” researchers found that the Tripod’s factor structure did not “hold up.” That is, Tripod’s 7Cs (i.e., seven constructs including: Care, Confer, Captivate, Clarify, Consolidate, Challenge, Classroom Management; see more information about the 7Cs here) and the 36 items that are positioned within each of the 7Cs did not fit the 7C framework as theorized by instrument developer(s).

Rather, using the MET database (N=1,049 middle school math class sections; N=25,423 students), researchers found that an alternative bi-factor structure (i.e., two versus seven constructs) best fit the Tripod items theoretically positioned otherwise. These two factors included (1) a general responsivity dimension that includes all items (more or less) unrelated to (2) a classroom management dimension that governs responses on items surrounding teachers’ classroom management. Researchers were unable to to distinguish across items seven separate dimensions.

Researchers also found that the two alternative factors noted — general responsivity and classroom management — were positively associated with teacher value-added scores. More specifically, results suggested that these two factors were positively and statistically significantly associated with teachers’ value-added measures based on state mathematics tests (standardized coefficients were .25 and .25, respectively), although for undisclosed reasons, results apparently suggested nothing about these two factors’ (cor)relationships with value-added estimates base on state English/language arts (ELA) tests. As per authors’ findings in the area of mathematics, prior researchers have also found low to moderate agreement between teacher ratings and student perception ratings; hence, this particular finding simply adds another source of convergent evidence.

Authors do give multiple reasons and plausible explanations as to why they found what they did that you all can read in more depth via the full article, linked to above and fully cited below. Authors also note that “It is unclear whether the original 7Cs that describe the Tripod instrument were intended to capture seven distinct dimensions on which students can reliably discriminate among teachers or whether the 7Cs were merely intended to be more heuristic domains that map out important aspects of teaching” (p. 1859); hence, this is also important to keep in mind given study findings.

As per study authors, and to their knowledge, “this study [was] the first to systematically investigate the multidimensionality of the Tripod student perception survey” (p. 1863).

Citation: Wallace, T. L., Kelcey, B., &  Ruzek, E. (2016). What can student perception surveys tell us about teaching? Empirically testing the underlying structure of the Tripod student perception survey.  American Educational Research Journal, 53(6), 1834–1868.
doiI:10.3102/0002831216671864 Retrieved from http://journals.sagepub.com/doi/pdf/10.3102/0002831216671864

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.