Actual Growth Measures Make A Big Difference When Measuring Growth

As a frequent critic of how states and districts present and interpret their annual testing results, I am also obliged (and indeed quite happy) to note when there is progress.

Recently, I happened to be browsing through New York City’s presentation of their 2014 testing results, and to my great surprise, on slide number four, I found proficiency rate changes between 2013 and 2014 among students who were in the sample in both years (which they call “matched changes”). As it turns out, last year, for the first time, New York State as a whole began publishing these "matched" year-to-year proficiency rate changes for all schools and districts. This is an excellent policy. As we’ve discussed here many times, NCLB-style proficiency rate changes, which compare overall rates of all students, many of whom are only in the tested sample in one of the years, are usually portrayed as “growth” or “progress.” They are not. They compare different groups of students, and, as we’ll see, this can have a substantial impact on the conclusions one reaches from the data. Limiting the sample to students who were tested in both years, though not perfect, at least permits one to measure actual growth per se, and provides a much better idea of whether students are progressing over time.

This is an encouraging sign that New York State is taking steps to improve the quality and interpretation of their testing data. And, just to prove that no good deed goes unpunished, let’s see what we can learn using the new “matched” data – specifically, by seeing how often the matched (longitudinal) and unmatched (cross-sectional) changes lead to different conclusions about student “growth” in schools.

The Great Proficiency Debate

A couple of weeks ago, Mike Petrilli of the Fordham Institute made the case that absolute proficiency rates should not be used as measures of school effectiveness, as they are heavily dependent on where students “start out” upon entry to the school. A few days later, Fordham president Checker Finn offered a defense of proficiency rates, noting that how much students know is substantively important, and associated with meaningful outcomes later in life.

They’re both correct. This is not a debate about whether proficiency rates are at all useful (by the way, I don't read Petrilli as saying that). It’s about how they should be used and how they should not.

Let’s keep this simple. Here is a quick, highly simplified list of how I would recommend interpreting and using absolute proficiency rates, and how I would avoid using them.

Under The Hood Of School Rating Systems

Recent events in Indiana and Florida have resulted in a great deal of attention to the new school rating systems that over 25 states are using to evaluate the performance of schools, often attaching high-stakes consequences and rewards to the results. We have published reviews of several states' systems here over the past couple of years (see our posts on the systems in Florida, Indiana, Colorado, New York City and Ohio, for example).

Virtually all of these systems rely heavily, if not entirely, on standardized test results, most commonly by combining two general types of test-based measures: absolute performance (or status) measures, or how highly students score on tests (e.g., proficiency rates); and growth measures, or how quickly students make progress (e.g., value-added scores). As discussed in previous posts, absolute performance measures are best seen as gauges of student performance, since they can’t account for the fact that students enter the schooling system at vastly different levels, whereas growth-oriented indicators can be viewed as more appropriate in attempts to gauge school performance per se, as they seek (albeit imperfectly) to control for students’ starting points (and other characteristics that are known to influence achievement levels) in order to isolate the impact of schools on testing performance.*

One interesting aspect of this distinction, which we have not discussed thoroughly here, is the idea/possibility that these two measures are “in conflict." Let me explain what I mean by that.

Five Recommendations For Reporting On (Or Just Interpreting) State Test Scores

From my experience, education reporters are smart, knowledgeable, and attentive to detail. That said, the bulk of the stories about testing data – in big cities and suburbs, in this year and in previous years – could be better.

Listen, I know it’s unreasonable to expect every reporter and editor to address every little detail when they try to write accessible copy about complicated issues, such as test data interpretation. Moreover, I fully acknowledge that some of the errors to which I object – such as calling proficiency rates “scores” – are well within tolerable limits, and that news stories need not interpret data in the same way as researchers. Nevertheless, no matter what you think about the role of test scores in our public discourse, it is in everyone’s interest that the coverage of them be reliable. And there are a few mostly easy suggestions that I think would help a great deal.

Below are five such recommendations. They are of course not meant to be an exhaustive list, but rather a quick compilation of points, all of which I’ve discussed in previous posts, and all of which might also be useful to non-journalists.

Large Political Stones, Methodological Glass Houses

Earlier this summer, the New York City Independent Budget Office (IBO) presented findings from a longitudinal analysis of NYC student performance. That is, they followed a cohort of over 45,000 students from third grade in 2005-06 through 2009-10 (though most results are 2005-06 to 2008-09, since the state changed its definition of proficiency in 2009-10).

The IBO then simply calculated the proportion of these students who improved, declined or stayed the same in terms of the state’s cutpoint-based categories (e.g., Level 1 ["below basic" in NCLB parlance], Level 2 [basic], Level 3 [proficient], Level 4 [advanced]), with additional breakdowns by subgroup and other variables.

The short version of the results is that almost two-thirds of these students remained constant in their performance level over this time period – for instance, students who scored at Level 2 (basic) in third grade in 2006 tended to stay at that level through 2009; students at the “proficient” level remained there, and so on. About 30 percent increased a category over that time (e.g., going from Level 1 to Level 2).

The response from the NYC Department of Education (NYCDOE) was somewhat remarkable. It takes a minute to explain why, so bear with me.

How Often Do Proficiency Rates And Average Scores Move In Different Directions?

New York State is set to release its annual testing data today. Throughout the state, and especially in New York City, we will hear a lot about changes in school and district proficiency rates. The rates themselves have advantages – they are easy to understand, comparable across grades and reflect a standards-based goal. But they also suffer severe weaknesses, such as their sensitivity to where the bar is set and the fact that proficiency rates and the actual scores upon which they’re based can paint very different pictures of student performance, both in a given year as well as over time. I’ve discussed this latter issue before in the NYC context (and elsewhere), but I’d like to revisit it quickly.

Proficiency rates can only tell you how many students scored above a certain line; they are completely uninformative as to how far above or below that line the scores might be. Consider a hypothetical example: A student who is rated as proficient in year one might make large gains in his or her score in year two, but this would not be reflected in the proficiency rate for his or her school – in both years, the student would just be coded as “proficient” (the same goes for large decreases that do not “cross the line”). As a result, across a group of students, the average score could go up or down while proficiency rates remained flat or moved in the opposite direction. Things are even messier when data are cross-sectional (as public data lmost always are), since you’re comparing two different groups of students (see this very recent NYC IBO report).

Let’s take a rough look at how frequently rates and scores diverge in New York City.

If Your Evidence Is Changes In Proficiency Rates, You Probably Don't Have Much Evidence

Education policymaking and debates are under constant threat from an improbable assailant: Short-term changes in cross-sectional proficiency rates.

The use of rate changes is still proliferating rapidly at all levels of our education system. These measures, which play an important role in the provisions of No Child Left Behind, are already prominent components of many states’ core accountability systems (e..g, California), while several others will be using some version of them in their new, high-stakes school/district “grading systems." New York State is awarding millions in competitive grants, with almost half the criteria based on rate changes. District consultants issue reports recommending widespread school closures and reconstitutions based on these measures. And, most recently, U.S. Secretary of Education Arne Duncan used proficiency rate increases as “preliminary evidence” supporting the School Improvement Grants program.

Meanwhile, on the public discourse front, district officials and other national leaders use rate changes to “prove” that their preferred reforms are working (or are needed), while their critics argue the opposite. Similarly, entire charter school sectors are judged, up or down, by whether their raw, unadjusted rates increase or decrease.

So, what’s the problem? In short, it’s that year-to-year changes in proficiency rates are not valid evidence of school or policy effects. These measures cannot do the job we’re having them do, even on a limited basis. This really has to stop.

Data-Driven Decisions, No Data

According to an article in yesterday’s Washington Post, the outcome of the upcoming D.C. mayoral primary may depend in large part on gains in students’ “test scores” since Mayor Adrian Fenty appointed Michelle Rhee to serve as chancellor of the D.C. Public Schools (DCPS).

That struck me as particularly interesting because, as far as I can tell, Michelle Rhee has never released any test scores to the public. Not an average test score for any grade level or for any of the district’s schools or any subgroup of its students. None.

A Below Basic Understanding Of Proficiency

Given our extreme reliance on test scores as measures of educational success and failure, I'm sorry I have to make this point: proficiency rates are not test scores, and changes in proficiency rates do not necessarily tell us much about changes in test scores.

Yet, for example, in the Washington Post editorial about the latest test results from the District of Columbia Public Schools, at no fewer than seven different points (in a 450 word piece) do they refer to proficiency rates (and changes in these rates) as "scores." This is only one example of many.

So, what's the problem?