Is StudentsFirst Conceding Permanent Achievement Gaps?

Or don't.

Or don’t.

I missed this when it first came out, but back in October StudentsFirst apparently released a report called A Personalized Future for Education. If you cut through the platitudes about “teaching with a 19th century model” designed for the needs of an “industrial-based economy”, the report’s core recommendations are not obviously unreasonable:

Personalized learning is a student-centered approach to education that allows each student to advance through academic content at his or her own pace. In a personalized model, also known as a competency-based education (CBE), time is the variable and learning is the constant, so a student’s competency is prioritized over his or her age. Personalized learning removes the one-size-fits-all approach to education by offering an array of choices and content to every student at a pace that meets his or her specific learning needs.

According to CompetencyWorks, a leading collaborative initiative that works to provide information and knowledge about CBE, there are five components of CBE:

1. Students advance upon mastery;
2. Competencies include explicit, measureable, transferable objectives that empower students;
3. Assessment is meaningful and a positive learning experience for students;
4. Students receive timely, differentiated support based on their individual needs; and
5. Learning outcomes emphasize competencies that include application and creation of knowledge along with the development of important skills and dispositions.

There’s nothing really revolutionary here; many of these things have gone on to one degree or another for a long time under other names (e.g., “tracking”). Strictly implementing this sort of “competency-based education” on a large scale would entail many substantial changes, but teachers, schools, and districts strive for imperfect approximations of these recommendations all the time.

Of course, the report is also peppered with StudentsFirst’s standard recommendations about things like choice for students and parents and accountability for teachers and schools. What’s really striking, though, is how uncomfortably the CBE proposals fit within StudentsFirst’s broader war against achievement gaps and “excuses”.

To see why I’m surprised to see StudentsFirst endorse the CBE model, it’s helpful to start with the group’s mission statement, which includes this:

Inside our schools, a great teacher is the single most important factor in a child’s education. While there are many factors that influence a student’s opportunity to learn, a great teacher can help any student overcome those barriers and realize their full potential. For this reason, we believe in doing everything we can to make sure teachers are supported and all schools are able to hire and retain the best teachers possible.

That’s a bit difficult to parse – just try to articulate StudentsFirst’s position on the importance of out-of-school “factors that influence a student’s opportunity to learn” – but I think it’s fair to say that this is the premise that motivates their side of the teacher quality/”poverty is not an excuse” debates. On StudentsFirst’s account it shouldn’t matter whether there are various, school-independent “barriers” to “a student’s opportunity to learn”: the point of education reform is to enable schools to “overcome those barriers”.

SF CBEAs far as I can tell, the StudentsFirst position has consistently been that if identifiable groups of students are consistently under-performing relative to their peers, that can be viewed as a sign of educational failure. After all, if teachers and schools are “overcoming” students’ various barriers, students should be progressing academically at relatively constant rates across populations.

Now let’s return to their competency-based education report, which includes this passage on the role of competency in education:

Grade-level promotion has historically been dictated by students’ age and attendance. This system of promotion takes a one-size-fits-all approach to education, and typically serves students in the middle of the class; students who need additional review end up being moved along before they’re ready, while students who have already achieved mastery on the given topic are prevented from moving ahead and often end up bored and disengaged. This model is problematic because it does not meet the individual needs of each student and may actually contribute to a gap in achievement between students.

The good news is that personalized learning provides educators with an alternative – an alternative that is already in high demand. Personalized learning allows students to learn at their own pace, advancing only when they achieve competency and mastery of the material. A student will spend as much time as they need to gain competency. If the student falls too far behind, educators can leverage data about where the student is and target additional supports to help him or her catch up. Additionally, personalized learning adapts to situations where a student is ahead in one subject and behind in another. This means that students are provided with an education that is custom-made for them based on academic needs, not age or seat time requirements completely unrelated to learning.

Emphasis added.

All of that sounds plausible enough to me, but as far as I can tell it completely cuts against (what I think of as) the standard StudentsFirst line, which is that the point of school reform is precisely to make sure that non-school factors aren’t slowing kids down academically. So I’m surprised to see StudentsFirst advocate a system that basically concedes – and builds around – the fact that kids don’t progress academically at the same rate.

Am I missing something here?

This entry was posted in Education Reform. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.


  1. EB
    Posted January 22, 2014 at 8:18 AM | Permalink

    It sound to me as if the Students First premise is that when learning is individualized, the students who start out behind are given a more powerful learning platform than in a typical classroom, where they are merely expected to somehow catch on to material that’s pitched a little (or a lot) ahead of where they’re starting. And if your definition of achievement gap is couched in terms of what per cent of each demographic group gets over the “pass” bar for each type of standardized test, this could (if true) result in higher pass rates for lower performing students and stable pass rates for higher performing students (who are all passing already). Voila! diminishing achievement gaps.


    • Posted January 22, 2014 at 12:25 PM | Permalink

      It’s conceivable that SF is intent on using proficiency gaps as opposed to average score gaps to describe achievement gaps, but I think SF is knowledgeable enough and sincere enough to know that that would be silly. Granted, that’s about the only way that I can square this report with their other stated positions, but it’s one I find hard to accept.

  2. Hainish
    Posted March 17, 2014 at 10:10 AM | Permalink

    The achievement gap is between identifiable demographic groups. SF doesn’t assume that personalized education will lead to differences among these groups. There’s no inherent contradiction there.

    Now, it’s possible that their model will lead to this outcome, but I don’t think it needs to. If students start out “behind” (lacking in background knowledge), they can catch up and move ahead (presumably better under a personalized model than by being moved along with a larger group).

    • Posted March 17, 2014 at 12:48 PM | Permalink

      It is logically conceivable that that’s how SF thinks they can thread the needle here, but that would require them to adopt the following peculiar combination of assumptions:

      1) Non-school factors affect the rate at which students progress (already admitted).
      2) Those non-school factors are randomly distributed and do not correlate with other demographic criteria.

      Maybe SF believe #2 as well as #1, but that strikes me as neither what they think nor what they would want to commit themselves to.

      As for students catching up, if some students are starting out behind, I’m not sure why we’d expect them to “catch up” (on average) in a personalized model. Why aren’t the “ahead” students continuing to make progress at at least the same rate?

      • Hainish
        Posted March 17, 2014 at 12:58 PM | Permalink

        I’m not sure why we’d expect them to “catch up” (on average) in a personalized model. Why aren’t the “ahead” students continuing to make progress at at least the same rate?

        You’re assuming that it takes the same amount of time to progress on all content. I was assuming that more-basic content can be progressed through more easily/quickly, allowing catching up. Also, I wasn’t assuming that all students would spend the same amount of time on academic work, but that those who start behind will spend somewhat more time on it (maybe not a whole lot more, but some).

        Regarding point 1: Yes, but the degree at which they do so can vary with what happens in school.

        Regarding point 2: Many non-school factors will not correlate, or will correlate only partially, with more-common demographic criteria. (I’m thinking of more innate factors here.)

        • Posted March 17, 2014 at 5:55 PM | Permalink

          It’s possible that kids can progress through more-basic content more rapidly, although it’s not obvious to me why we’d expect that to happen on average. In any case, even if we assume the lower-achieving kids can progress through the basic stuff more rapidly than advanced students can make progress, that isn’t enough to close achievement gaps, just narrow them, because however far along the lower-scoring kids get, the more advanced kids will have been at that point at some time in the past and (presumably) have made progress since then. A gap therefore would remain.

          Re: point 2, note that for SF’s position to *not* imply permanent gaps it would need to be the case that non-school factors taken together do not correlate *at all* with common demographic criteria. That is the assumption I find so questionable that I doubt SF wants to explicitly advance it (and I think I could find examples of them explicitly or implicitly denying it in the past, such as when they acknowledge that “poverty matters”, etc.)

Leave a Reply