20 Minutes with Peter Liljendahl - Assessment

*WARNING*  The following post is just a bunch of random thoughts...and isn't well worded.  It's just a jumble of thoughts and is supposed to help me understand the brief phone interview I had with Dr. Peter Liljendahl**

In last week's presentation Dr. Peter Liljendahl alluded to much of his work on assessment but but given time constraints he couldn't really talk about it.  I was so enthralled by Peter's work that I contacted him afterwards and requested to talk to him.

So I got 20 minutes today with Dr. Peter Liljendahl and I wanted to pick his brain about assessment as much as possible.  He has done much work with assessment, but is still in the process of formalizing it in the form of a paper.

Since it's Mother's day - I can't do a great job of writing it out - but I feel that it's just important for myself to just get it off my chest so I can process it quicker.

Formative assessment, according to Peter's research, is much more effective when it informs students of where they are and where they are going.  Getting a mark of 13/20, where all the marks are aggregated, provides a nice overall picture but doesn't inform the student of the specifics which can help improve and move them forward.   He has found that 50% of students improve by a letter grade when they are specifically informed of where they are and where they are going.  The other 50% are already performing at maximum, or do not want to improve, or something else.

It turns out that teachers have a 'code' in their head about questions.  In a study, Peter gave students and teachers five questions and asked them to put them in 'order' of complexity.  The students had no idea, but the teachers could do it.  If the students have a better understanding of this 'code' that teachers have, they improve their marks.

One teacher has been playing around with communicating the "code" in our head by giving out something like this:

I have done something similar - but I used checkboxes instead of a continuum.  The continuum is an important touch as it will allow for better communication for themselves and teacher.  This allows for feedback, and students must do it during the class time at a semi-regular basis (daily is too much).

Another idea regarding evaluation, and 'reporting out' is shown below.  Again, this way of parsing out the information allows students to take their time in learning and make explicit what they need to be working on.

One teacher gave out one sheet at a time - level 1 questions on one sheet.  Level 2 and 3 questions on another sheet.  It looks like in each case, if one can achieve level 3 at a consistent level for a concept, I would award that person their full marks - regardless if they failed level 1 or 2 at an earlier stage.  Sometimes it takes students a while to understand the first two levels of a question before 'getting it'.  This will work nicely with the spiral curriculum...but the tracking seems like a nightmare to me at the moment.  I'm also double thinking this....shouldn't those who spend the time to understand each day, be rewarded for understanding 'faster' than the next person?  Perhaps not - math isn't really about speed, now is it.  Deep understanding IS the more important thing I would want to emphasize.  Fast, speedy math will more likely encourage memorization - which is what computers should be doing...not us humans.

Apparently, this type of reporting increases student autonomy as students.  If I were to allow students to report out an understanding of concepts through not just quizzes or tests...but also through homework or through the on the board work and have them complete a portfolio of 'proof'.

I am not sure about that...I'd really have to work on their mindset first...to really ONLY care about learning.  This reporting out would have to be...somewhat controlled by me or somehow checked to make sure for accuracy.

I really don't know what to make of all this.  I think I need more processing time on all of this...and have many questions.