Pages

Sunday, January 21, 2018

Something new for 2018: "From L to J"

Last October, our district hired Lee Jenkins to come and train our grades 4-12 staff on a school improvement strategy he calls "From L to J".  I had never heard of Lee Jenkins before, but I left that day knowing that I wanted to integrate "From L to J" into my classroom.

In case you're not aware of what "From L to J" is, here is a very brief overview:

  • Teachers write learning targets for the entire year before the year begins; the learning targets for the year are given to students on the first day of school.
  • Teachers write an assessment question aligned to each learning target.  Questions are typically entered into a PowerPoint / Slides presentation.  This is done at the start of year.
  • Each week, students are given an "L to J" quiz.  Quiz questions are randomly selected each week.  The number of questions is a function of the number of learning targets.

A few more details:
  • Each quiz, students track their own progress by plotting the number of questions answered correctly into a histogram.
  • Early on, students may get very few questions correct.  This is to be expected if the learning targets are things that students haven't previously learned.
  • Yes, quiz questions may be over concepts and topics that haven't yet been covered in class.  Because of this, "From L to J" quizzes do not effect a student's grade.  
  • As the year progresses, [theoretically] students will begin to get more and more questions correct.
  • After each quiz, a class total of number of correct answers is calculated and plotted.  Each time the class total reaches an all-time best, there is a "celebration".
  • [If you're wondering where the term "From L to J" originates, it comes from creating a class histogram after each quiz.  On the x-axis is the number of questions answered correctly and on the y-axis is the number of students.  Each student is a data point in the histogram.  Early on, many students should get 0, 1, or 2 questions correct, creating an "L" shape distribution.  Over time, the histogram begins to take a bell-curve shape.  And by the end of the year, the histogram is hopefully shaped like a "J", meaning a lot of students got most or all of the questions correct.  Take a look at some examples here.]

Jane, Jarrod, and I are the three geometry teachers in our high school.  We spent a day writing our learning targets and creating quiz questions.  We have 72 learning targets and decided to create three different questions for each target.  [As of right now, we have one question for each target and are working on completing the other questions.]

Because we are implementing this at the beginning of the second semester, our histograms shouldn't necessarily ever be shaped like an "L".  Next year, when we implement this starting at the beginning of the year, I expect a much truer "L to J" transition.


What was I drawn to with this strategy?
  • In his presentation, Mr. Jenkins talked a lot about how this idea holds students accountable to remember what they have learned.  Too often students believe that once they take the test over a topic or concept, they can forget about it.  Or maybe it's that we [as teachers] allow them to forget about it.  The "From L to J" quizzes provide a systematic spiral review for students.  There is randomness in which learning targets are reviewed, but my feeling was that this review of previous topics is better than no review.
  • Students are aware of exactly what they are expected to learn (and retain) throughout the year.  We have already been using learning targets with the students since the start of the year; this strategy provides a bit more formality to the learning target goals.
  • Mr. Jenkins also talked a lot about how this strategy helps students who have a tendency to struggle.  Even if a student is getting a D or F for a grade, I can point out to these students that they are still learning something.  [Future potential: Standards-based grading!?!]
  • Mr. Jenkins talked about how this strategy scored high on John Hattie's effect size research.
  • There is a little bit of statistics that gets worked into our classes.

What were my hesitations about implementing this strategy?
  • Time.  We're assuming that the quiz each week will take about 15 minutes.  That's 15 minutes a week that I won't have to do other class activities.
  • Quiz questions need to be DOK level 1 & 2.  The questions need to be able to be answered fairly quickly (< 90 seconds per question) and have one concrete answer.  I can't ask questions where students are asked to explain their thinking because it would be too hard to score.  My hands are tied with recall / skill level questions.

Last week, we took our first "From L to J" quiz.  It seemed to go pretty well with the students.  I found the randomness of the questions fun; my first class of students drew a lot of review questions and I had a lot of high score because of it.  My second and third classes had at least two questions that were over topics that we haven't yet covered.  Needless to say, they didn't score as well.  

We will have another one on Tuesday of this week.  I may post an update midway through the semester and tweet highlights and no-so highlights along the way.  I'm excited to try this out and I'm looking forward to doing it for the full year next year.