Monday, December 16, 2019

What Is A Day Of Learning, Anyway?

The measure crops up frequently in discussions of education policies and, sometimes, products. But what the heck does it even mean?
Charter advocates like to point to a CREDO study that shows urban charters giving students an additional 40 days of learning growth in math and 38 in reading (while critics bring up the 2013 CREDO study finding that charter schools provided seven additional days of learning per year in reading and no significant difference in math). IndianapolisNew York City, and other big systems find charter advocates touting additional days of learning. 
Meanwhile, one of the widespread criticisms of online schools is a CREDO study which found that cyberschool students lost 72 days of learning in reading and a whopping 180 days in math–that’s a whole year.
Bridge International Academy describes its success in Kenya in terms of added days of learning. Research into the educational effects of variables such as teacher experience is expressed in days of learning. Sales representatives for edu-products will promise additional days of learning.
But what exactly is a day of learning? Classroom teachers know that a Monday is not equal to a Friday or a Wednesday. Surely it’s not the day that students get out early, or the day that is interrupted by an assembly, or the day that the teacher was pulled out for meetings, or the day that the baseball team was dismissed early for an away game. Certainly not the day that everyone in school was reeling and preoccupied because of a local tragedy. A day in September is not the same as a day in April, and certainly not any day in the season that we’re approaching, because from mid-November until the end-of-year break classroom teachers are extra-challenged to get a day out of a day. 
So when is it? When does this proto-typical day, this day on which exactly one day’s worth of learning occurs? Where is education’s answer to Lebanon, Kansas (the geographic center of the contiguous U.S.)? Is it a statistical anomaly like the1.9 children being raised by the average U.S. family? Can this measure be broken down more precisely? Can we talk about hours of learning? Minutes? Seconds?
The Learning Policy Institute offers an explanation for days of learning. The short form is that a typical growth on a standardized test score, divided by 180, equals one day of learning. If you want a fancier explanation, LPI looks via CREDO to a 2012 paper by Erik Hanushek, Paul Peterson, and Ludger Woessmann:
To create this benchmark, CREDO adopted the assumption put forth by Hanushek, Peterson, and Woessman (2012) that “[o]n most measures of student performance, student growth is typically about 1 full standard deviation on standardized tests between 4th and 8th grade, or about 25 percent of a standard deviation from one grade to the next.” Therefore, assuming an average school year includes 180 days of schooling, each day of schooling represents approximately 0.0013 standard deviations of student growth.
So in the end, “days of learning” has nothing to do with days or with learning. It’s simply another way to say “this policy or product seems to correlate with an increase or decrease of scores on a standardized test of reading and math.” 
Learning can’t be measured in days or minutes or inches or pounds or hectares. Pretending that you can use test scores, assumptions and standard deviations to measure learning the same way you can portion out milk in a measuring cup is not science–it’s rhetorical smoke and mirrors.
If you wonder why classroom teachers are not more engaged with or moved by educational research, here’s one reason–because the euphemisms and constructs of researchers use a frame of reference totally removed from the experience of classroom teachers, designed to hide what they’re really talking about instead of illuminating it. Someone who approaches a classroom teacher and says, “I’ve got a way for you to get more days of learning out of your students,” should not expect to be taken seriously.


  1. How about focusing on a real issue: chronic absenteeism (de-facto truancy). In so-called "failing schools" it is not uncommon for students to miss 20 to 30 to 40+ days of school. At the secondary level, missing one day of school is academically debilitating for marginal students. They have missed instruction, and missed work which often relies on the missed instruction is piled on top of current work. Missing 5 to 10+ days in a marking period is insurmountable. Missing time also produces frustration that often leads to behavior issues. Now their suspended. More time out of class. There are hundreds of thousands of students who fit this description yet these no-thing edu-meddlers want to waste time with bogus algorithms.

  2. "Educational research" is an oxymoron...

  3. There is a hunger from teachers using social media to find research which does tell them what to do...

  4. You are over-working it.

    We use lots of measures of things that aren't exactly equal, but even out in the long run. Not every home run or touchdown is equal -- some are game winning and some aren't. Some are in miserable weather, and some aren't etc. But that doesn't make the measure worthless.

    If my boss asks me how long it will take to do a job, I can't answer "but every day is not the same" because he will think I'm an idiot. I tell him "about five days" and he knows how long it will take (and because he's not an idiot, he knows it's not exactly five every time).

    "Days of Learning" works for me. Grade increases don't, because they will differ according to starting point.

    If we want to improve our teaching, how are we *meant* to measure which way works best? Personal evaluation sure doesn't -- I've known teachers swear by methods that were total rubbish.

    Other fields have issues with measurement -- medicine, since not everyone is the same. Yet they work around those issues, and they make progress.

    Refusing to measure just makes someone look like they're not interested in any reform at all. Which makes sense only if your current system is perfect.