Round 6.1: Concept Maps

Aug 18, 2013 • Greg Wilson

Our sixth cohort had their first meeting on August 15; more than two dozen people attended the two meetings combined from as far away as Poland, Singapore, Brazil, New Zealand, and Oklahoma. We discussed the goals for the course, and set the stage by exploring the key cognitive distinctions between novices, competent practitioners, and experts (see below the fold for a complete transcript of the Etherpad session).

For the first practical exercise, please do the following:

  1. Read the chapter by Mark Guzdial on why teaching programming is hard and the first two chapters of How Learning Works.
  2. Pick a small topic—something you could teach in 10 minutes or less—and create a concept map for it. Please pick something that is related to Software Carpentry (i.e., something to do with computing skills and scientific research), and please don’t worry about making the concept map look pretty: as we discussed, you’ll get more honest feedback on a sketch than on a work of art.
  3. Post your concept map to the blog (in the category “Round 6.1″), and then comment on at least two other concept maps. Are they comprehensible? Are they too large/too small? Are there missing/superfluous bubbles and links?

Step #3 is actually the most important one. The concept maps are useful in their own right, but what we’re really trying to do is foster a culture of feedback and improvement.

Our next meeting(s) will be 11:00 and 19:00 on Thursday, August 29, on the same Etherpad and conference call number. Please try to have your concept maps posted no later than August 21, so that people have a full week to explore them and give feedback before our meeting.

Please add your name and location below (note: this is always a good demo of why concurrent/parallel programming is hard :-)

Overall theme: educational psychology instead of the actual two-day content.

Theme 1. We can use web tools ourselves during our learning how to teach
Eg Etherpad
We use Etherpad in class for note-taking — advanced learners who might  otherwise be bored can do this instead of reading facebook during the  lesson
Do it during our meetings to get some practice

Theme 2. Novice vs competent practitioner vs expert

Novices don’t know what they don’t know. They ask questions that don’t make any sense. They have taken the knowledge and put it into the mental boxes that they already have. They don’t have a conceptual map of the knowledge domain.

Our goal is to create those mental boxes, provide novices with map.

Competent practitioner: more or less correct map or the problem, can solve problems in a reasonable amount of time with reasonable amount of success
— if we get people here, we can consider it a success (want to get people from novice to here, not from competent -> expert)

Experts: know when to break the rules (e.g. for special cases), can diagnose and reverse engineer the problem better than competent practitioners
— You don’t have to be an expert in a domain to teach, actually it is better if not (experts are bad at teaching novices because they have forgotten what it is like to not understand something — called expert blind spot)
— Experts make things that are actually difficult until hours of practice, sound too easy -> very demoralizing to novices

An expert knows when to break the rules, and has a deep enough understanding to know when to do something different. When not to follow the procedure. Can identify experts as they are much better at diagnosis and debugging. Competent practitioners can work forward. Experts can work backwards from the symptoms to find the likely causes.

Learning timescales
Our goal is not to bring people from Competent to Expert; we can’t do that. That only happens via practice and reflection. This takes a lot of time: 100s to 1000s of hours.

The time it takes to move from Novice to Competent is much smaller: 10s to 100s of hours.
Time to go from novice to competent is critically dependent on the quality of the instruction.

Research shows that the only way from competent to expert is practice and lots of reflection (reflection is most important). Need to take the time to think about what you did. Time involved is thousands of hours.

Content changes so we focus on education/teaching.
Novice -> Competent practitioner -> Expert
Goal: Take students from from Novice (no mental map) to Competent practitioner (someone who has a mental map and can work forward).
Experts are often bad at teaching novices! Remembering what is like to not know is key.
Concept map: 

  • Key concepts (design the lesson), how are they related. Order in which ideas are introduced — dependencies!. Things depend only on things you’ve already introduced.
  • Can be also useful for the students: hand out or write (on board) as you go through the lesson.
  • Students can also do their own concept map — very good feedback.

Continuous feedback is very important.
*Remember the relationship between how linguistic and visual memory works. Concept maps are a great way to teach and obtain feedback
By Wed. post your own concept map on a topic that can be explained in 10 minutes.Comment on at least two others’. Rough drawing is deliberate: the more polished it looks, less quality feedback you get.

One block: Talk for 10 min, practice for 5 min. Short-term memory!

(not all the way to expert)

To get to expert: Time involved is ten to thousands of hours.

Most instructors are competent (few are experts).  We are actually better off NOT being experts, illustrated by the term:

Expert blind spot
Stereotype threat

Expert aren’t good at teaching novices.
Experts have forgotten what it is like not to know something.
The best person to teach novice is someone who has been novice recently, because they remember what it is like to be novice.  It is a better experience for the novice.
Experts can make it seem easy, when it is not for the novice.

Greg’s promise:
Make us not-bad teachers

Exercise 1 by Wednesday, August, 21: Something you can teach in 10 minutes or less that falls inside the bounds of software carpentry and is in the interest of scientsts. Draw a concept map (see link on line 33 for examples). If concept map has more than a handful of links than you are trying to cover too much. (Is there a general rule for # of blobs => # of minutes?)

Concept map — half a dozen bubbles per lesson as rule of thumb.

Complete by Wednesday of next week (8/21/2013)

Look at others (at least two) and comment on theirs, and look at their comments on yours. This will let you practice reflective thinking.

Every time you use the work ‘later’ you lose half of your learners.
Forward references = extra cognitive burden
These are used three ways: (1) Some people hand out concept map at outset of lecture. (2) Others draw map as the lesson progresses; they can see how they got where they are. (3) Have learners draw concept map to show what they think they just learned. You can then see if they are understanding this (labor intensive, small scale).

How to give and get feedback. If you give constructive feedback often enough, you can step away from your performance and correct problems. Can learn how to listen to yourself. Key factor on how well someone teaches, is can a teacher evaluate their own teaching performance (give themselves feedback)?  US K-12 teachers evaluated for ~ 4 hours/year by peers? vs Finnish teachers are evaluated by another teacher for 3 hours a week. Evaluation to say ‘this went well, this didn’t’. Finnish education system excels. Everyone has a culture of continious feedback — how to give it and receive it. Feedback is absent in the anglo-saxon education system. In this course we will learn the art of feedback — how to give and receive. Giving feedback will likely take longer than creating the concept map.

Why rough prototypes? The more polished something looks, the less honest the feedback you recieve will be — Jeb (sp?) Johnson et al

If a client sees a UI designer put it together in 30 secs in pen and crayon, then they do not worry about giving feedback (won’t be worried about hurting their feelings — will not pull any punches)

Why <=10 minutes?
Reason for 10 minutes: your short term working memory only holds so much. Go for ~6 things and then give people time to practice so they digest.
Also, human brains can only pay attention for 45-90 minutes before fatigue. This is a physical constraint. (aka if you continue for more than 45 minutes, you hit the mmblr mmblr  wall)
You cover less material this way, but the learners can remember things this way.

=============================
TO DO FOR NEXT TIME

Exercise 1 by Wednesday, August, 21: Something you can teach in 10 minutes or less that falls inside the bounds of software carpentry and is in the interest of scientsts. Draw a concept map (see link on line 33 for examples). If concept map has more than a handful of links than you are trying to cover too much. (Is there a general rule for # of blobs => # of minutes?)

Complete by Wednesday of next week (8/21/2013)

Look at others (at least two) and comment on theirs, and look at their comments on yours. This will let you practice reflective thinking.

Think about what using etherpad to manage discussion is like
—> And how we can use it better

Please read:

Reason to meet every two weeks: do assignment in first week, get/give feedback in second week.