Round 8.1/8.2: From Concept Maps to MCQs

Jan 31, 2014 • Greg Wilson

Meeting of the Software Carpentry Instructors Study Group

Round 8.1/8.2 (2014-01-30)

For Feb 13

  • Pick a topic you could teach in 10 minutes (can be the same as the one used for exercise 1, or different if you’d rather)
  • Create and post:
    • One multiple-choice question to distinguish novices from competent practitioners (Categories “8.2” and “Assessment”) before you start teaching
      • 4 answers, all are plausible
      • wrong answers should be “distractors”
      • Questions should not be “self assessment” style questions
        • The less folks know about something, the less accurately they were able to assess their ability. (Dunning-Kruger effect)
      • Eg. which of the following pipelines will display the following text
    • One multiple-choice question to tell whether your audience has actually learned what you set out to teach
    • One short practical exercise they could do to exercise their new knowledge
      • Better to write something to fill in/fix code rather than asking to start from scratch
        • both for the students and for you marking it
    • Do not include answers to your MCQs or exercise
  • Comment on at least two other people’s postings
    • Answer their MCQs! Do the exercise! Post your solutions as comments!
  • Read Chapter 3 from How Learning Works on motivation
    • Participants in Group 7 are posting motivational/demotivational stories in the coming week — once you’ve read Chapter 3, please do comment on them as well

Notes

7:00 pm

  • Greg Wilson: no concept map
    • What did you learn about your topic from doing this?
    • What did you learn about how you think?
    • What did you learn about making concept maps (and are they useful)?
  • James Harmon: http://teaching.software-carpentry.org/2014/01/30/what-is-a-dag/
    • There are lots of relationships between the concepts in the topic I chose (I could have drawn more but the diagram was getting too confusing)
    • I don’t know as much about the relationship and organization of my knowledge as I might have thought
    • Concept maps are not necessarily hierarchical
  • Brenna O’Brien: http://teaching.software-carpentry.org/2014/01/21/concept-map-objects-in-javascript/
    • objects in JS are a complex topic, need to oversimplify for a beginner lesson
    • I naturally think about things like this, though it usually comes *after* expertise
    • seems like a good way to keep scope of a lesson limited, so yes
  • Likit Preeyanon: http://teaching.software-carpentry.org/2014/01/22/basic-text-processing-with-python-built-in-string-methods-with-examples/
    • It isn’t easy to make a concept map for some topics.
    • There are many useful built-in methods for manipulating strings in Python and it is not difficult to learn how to use them.
    • I learned that I didn’t usually think of things in an organized way.
    • Making a map helps me find out what I don’t know — the missing link.
  • Atul Varma: http://teaching.software-carpentry.org/2014/01/21/concept-map-how-git-works/
    • Git is still too complicated, but a bit less complicated than i thought it was, maybe
    • I like making connections between big concepts.
    • The process of drawing them out definitely helped me figure out what my own knowledge organization was and how everything fit together. Also, cycles in concept maps usually describe concepts that make me say “whoa”.
  • Matt Hall: http://teaching.software-carpentry.org/2014/01/22/concept-map-find-and-replace-in-python/
    • I hadn’t thought carefully enough before about the differences between regex and parsers like lxml/BeautifulSoup (lesson came from Likit’s comment)
    • I was surprised at how hard I found it to ‘conceptify’ the topics I had chosen. Pretty sure I didn’t do the greatest job.
  •  Daniel Chen: http://teaching.software-carpentry.org/2014/01/16/concept-map-python-variables-and-syntax/
    • I learned that I was inadvertently thinking and changing between computer science-speak and normal-people speak (i.e. = as assignment and = as equality, when in csci == is equality)
    • After speaking with some of the professors here (aron and randy) I still remember how it feels to be a novice!
    • They are useful for thinking and remembring a topic.  I remember commenting on a SQL map, how the where clause works and how it extends to nesded select queries.  For more complicated topics, they are very good for covering all pre-requisite topics before continuting to a more advanced topic
  • Martin Paulo: http://teaching.software-carpentry.org/2014/01/22/concept-map-python-variable-scope/
    • Topic: That even what I view as a fairly simple act has a layer of complexity that could trip novices up.
    • Thought: That there’s a whole lot of abstracting going on behind the scenes :)
    • Concept maps: That I have a knee jerk reaction to create flow charts: which concept maps definitely are not — and that it can be difficult to leave the flow charts behind :( And yes, I think that they are very useful: in my case they surfaced the fact that I was abstracting stuff. And reviewing other’s maps did give me some insight into their knowledge/thoughts.
  • Isabel Fenton: http://teaching.software-carpentry.org/2014/01/22/concept-map-loops/
    • Reading other peoples comments helped me to understand which parts of the topic I take for granted.
    • I don’t naturally think like that, but I found that creating one forced me to think more logically about the topic,
    • so I think it was useful, especially if I want to explain something to someone else, but also for myself
  • Zahra : http://teaching.software-carpentry.org/2014/01/29/creating-a-trace-table/
    • I learned sometimes even with simple concepts, it is hard to explain the concept by itself. When you know something, it is not easy to look at the concept from others point of view. I need to see the big picture to generate a concept map.
    • I jumped into explaning how to generate a trace table and not the concept map !
  • Brad Taber-Thomas: http://teaching.software-carpentry.org/2014/01/23/concept-map-batch-data-processing-script/
    • I learned that the actual steps in a batch script are probably less important to understand than the goal of the script
    • I learned that I tend to think from a very broad perspective (e.g., “goal” and “doc” were the first bubbles that came to mind when I thought about batch processing)
    • I learned the value of the act of making a concept map
  • Neem Serra: http://teaching.software-carpentry.org/2014/01/28/concept-map-unit-tests/
    • My topic was unit tests.  I learned how to break them down simply.
    • I learned that I’m good at flow diagrams but not concept maps!  I think I could be better at connecting the ideas behind my topic
    • I learned that I think in logical steps as in this thing leads to this thing.
    • I learned that concept maps were harder to write than I expected. I like making steps, but that doesn’t necessarily mean that I’m linking together concepts.  I also couldn’t remember how to concisely say the basic steps because I just know how to do it, but I’m not terribly good at explaining it out loud.
  • Evan Bianco: http://teaching.software-carpentry.org/2014/01/23/concept-map-using-git/
    • Topic: I think did more of a flow diagram than a concept map. More “how to use” rather than characterizing “Git” as a concept.
    • Thinking: I learned that I seem to place non-spatial concepts in a spatial way.
    • Concept maps: I use examples, analogies, and pictures to explain things (which could be dangerous). Comments from others were helpful.
  • Sheldon McKay: http://teaching.software-carpentry.org/2014/01/23/concept-map-conditional-statements/
    • I learned that even a simple-seeming topic takes more explaining than I thought.
    • My thinking steers me more towards flow charts than concept maps.  I’m still figuring out what a concept map is.
    • I learned that it took more bubbles than I expected to express what I was thinking.  Putting pen to paper makes it harder to take mental shortcuts.
  • Andrew Zonca: http://teaching.software-carpentry.org/2014/01/22/concept-map-git-branching-workflow/
    • a cleaner way to approach git branching (found reading different tutorials)
    • I find it difficult to find interconnections, I tend to simplify knowledge trying to organizing it as a list (…)
    • It was the first time I ever did a concept map, I find it very very difficult, however probably I chose not a suitable topic, and I was more concerned about drawing the workflow itself instead of the concept map required to explain it to somebody else (thanks to commenters!)

2:00 pm

  • Greg Wilson: no concept map
    • What did you learn about your topic from doing this exercise?
    • What did you learn about your thinking?
    • What did you learn about concept maps? Did you find them useful?
  • Abigail Cabunoc: http://teaching.software-carpentry.org/2014/01/22/concept-map-select/
    • Topic: I realized how central certain concepts (like fields and rows) are for different SQL concepts
    • Thinking: I learned that I’m a pretty graphical thinker.
    • Concept maps: it forced me to think about the relationships that exist in my mind between concepts. I often assume others will have the same relationships, so this helped to explicitly write them down.
  • Helen Yezerets: http://teaching.software-carpentry.org/2014/01/25/concept-map-java-object-class-variables-and-methods/
    • Learned: Helped me to structure      my own understanding
    • Visual structure helps to      concentrate on major ideas
    • Concept maps: excellent tool. I find it better than taking regular notes. Had to go over the structures multiple times
  • Chris Friedline: http://teaching.software-carpentry.org/2014/01/22/concept-map-ipython-notebook-for-documented-and-reproducible-computing/
    • topic: that it’s surprisingly difficult to summarize a concept that I use every day and that there are many connections among the bubbles in the map that I simply just take for granted (like public-key ssh)
    • thinking: that it wasn’t as difficult as i thought it would be to think visually in this way
    • concept maps: super useful, but really easy to get well out of scope into complex connections that may/may not lose the overall message
  • Jess Hamrick: http://teaching.software-carpentry.org/2014/01/19/concept-map-git-commit/
    • What I learned: git commands like “checkout” and “add” are consistent in terms of internal git structure, but *really* unintuitive to explain because at a high level, these commands have multiple behaviors.  I think I worked out a clearer idea of the interaction between HEAD and the index and the stage, etc. Still think I am a little confused, though.
    • What I learned about my thinking: It’s kind of piecemeal. Lots of connections between things, but those connections are not always well specified.
    • What I learned about concept maps: they can get really complicated really quickly, even if you’re trying to explain something that (should) be relatively simple.
  • Mark Stacy: http://teaching.software-carpentry.org/2014/01/23/concept-map-python-file-object/
    • What I learned: Keep audience in mind. What needs to be explained and not skipping subject matter.
    • My thinking: I tend to assume prior knowledge and get ahead of myself.
    • Concept Map: Focus my thoughts on what needed to taught and organized
  • Matthias Bussonnier  http://teaching.software-carpentry.org/2014/01/24/concept-map-image-embeded-in-ipython-notebook/
    • I was before saying unprecise things that ‘Embeded’ would not need internet, but it does at the time of generated the notebook.
    • Often difficult to categorize the different kinds of conccept/object you want to represent
    • Often difficult to have consistent meaning for the “arrow”.
    • I tend to want to write more a “Flow Chart” than a mind map.
    • I have more difficulties to layout things on paper that the way I “see” it in my mind.
  • Rob Beagrie: http://teaching.software-carpentry.org/2014/01/18/concept-map-what-is-a-program/
    • What I learned:
      • Although the concept of a program is very simple, it is very difficult to describe in terms that someone at a software carpentry bootcamp would find helpful…
    • My thinking:
      • I think I found that my thinking was really centred around problems — what is it that the programmer actually wants the computer to do?
    • Concept maps:
      • Turned out to be not as clear as I expected what should be a bubble and what should be an arc
      • My final version was very different to my first one… it definitely changed the way I would have taught this as a lesson
  • Alexis Pyrkosz http://teaching.software-carpentry.org/2014/01/27/concept-map-round-8-1/
    • Topic: I didn’t have as fixed/parallel a structure for the information as I’d originally thought
    • Thinking: I prefer outlines for teaching/graphics for research
    • Concept maps:Keep it simple because each circle could take a while to discuss
  • Patrick Marsh (Was sick with flu and didn’t complete new concept map)
  • David Schryer http://teaching.software-carpentry.org/2014/01/22/becoming-an-expert-concept-map/
    • I wanted a way to both capture my ideas on the subject and convey them to learners.
    • It was hard for me to decide what to add to the diagram and what to leave out.
    • The end result was a concept map with fallback solutions which turned it into a flowchart of sorts.

10:00 am

  • Greg Wilson: no concept map
    • What did you learn about your topic from doing this exercise?
    • What did you learn about your thinking from doing this exercise?
    • What did you learn about making concept maps? In particular, do you find them useful?
  • Melanie Segado: http://teaching.software-carpentry.org/2014/01/22/concept-map-secure-vnc-from-android-phone/
    • I had a conceptual flaw in how I thought about SSH tunneling
    • I tend to think in systems (Ie. how to get something TO work) as opposed to concepts (ie. why something works)
    • It’s very difficult to do them quickly, without colour coding. It’s also very difficult to keep them from turning into a flow chart (which mine sort of did, I think). Yes, I find them useful.
  • Gabriel Devenyi http://teaching.software-carpentry.org/2014/01/22/concept-map-text-processing-in-bash/
    • Text flow in Unix could easily be very confusing for a new user given the multitude of paths text can take
    • I think about text as a linear path on the command line, but it may be thought of better as a flow
    • Some topics which seem small quickly become very large concept maps. It was useful in re-structuring my idea of text in Unix.
  • Martin Callaghan: http://teaching.software-carpentry.org/2014/01/22/5608/
    • As far as the topic goes, I realised that I needed to get a clearer understanding of how the individual concepts are related. It’s one thing to have them in mind, entirely another to vocalise or visualise them.
    • Thinking wise, it was quite difficult to clearly summarise the concepts into six distinct points.
    • Concepts maps are quite useful and it was interesting to see how other people used them. Normally, I find that I tend to plan in linear lists rather than in 2d though.
  • Benjamin Bradshaw: http://teaching.software-carpentry.org/2014/01/22/concept-map-system-calls/
    • I realized the concept of system calls quickly gets into hairy low-level descriptions of x86 specific architecture when trying to explain them thoroughly.
    • I learned it is particularly difficult for me to realize the best “starting point” to explain an advanced topic to someone with no background in that topic.
    • I did find the concept map exercise useful.  In particular being able to get quick feedback from others on how I would go about teaching, before actually committing to a lesson plan.
  • Joon Ro: http://teaching.software-carpentry.org/2014/01/22/concept-map-git-how-to-contribute-to-an-upstream-with-a-feature-branch/
    • What I learn about
      • Some flows were hard to represent with arrows
      • Hard to find where to draw a line;  given I wanted to limit the number of circles.
      • Need to work on my control of articles
    • How I think
      • Major things first and then details; sometimes I forget about a detail and had to move major part around to make a room for the detail
      • Sailent of things I struggle to understand
  • Jeramia Ory: http://teaching.software-carpentry.org/2014/01/22/concept-map-unix-find-command/
    • Telling people which command to run is easy, explaining how it works much harder
    • Distilling something into a half dozen concepts is difficult. Unpacking your own assumptions and background takes time.
    • Many people have different ideas how they work. Some are more flow diagram, some more list like. I’m still not sure if I’m sold, but maybe I’m making them too difficult
  • Anne Moroney: http://teaching.software-carpentry.org/2014/01/22/concept-map-github-contribute-to-an-upstream-on-a-patch-branch/
    • Saw some gaps in my thinking/learning, not enough expert to teach, so added info, did research.
    • After doing it all, read the book [it arrived late, sorry!] and saw the hierarchical vs grid type of ‘advanced’ maps. I find the later totally unhelpful! Maybe they help the author, and if simple perhaps they help others, but I do not grok the value. Help??
    • I wrote an article on what I call ‘circle outlines’ years ago — I’m a huge fan.
  • Stephen Turner. Concept map for `find | parallel`: http://teaching.software-carpentry.org/2014/01/16/concept-map-for-gnu-parallel/
    • Topic — if was going to teach in more detail, would need to learn more about the syntax, why/how parallel works how it does. May also want to discuss installation.
    • Thinking — as a competent/expert user, sometimes would be easy to assume the novice knows something they don’t. Also, had tendency to turn concept map into flowchart.
  • Brian Miles: http://teaching.software-carpentry.org/2014/01/29/what-are-regular-expressions/
    • I may not know it well enough to teach it(?)
    • My thinking was too expert-oriented
    • Concept maps are a useful exercise for gauging my own understanding
  • JC Leyder: http://teaching.software-carpentry.org/2014/01/17/concept-map-yield-in-python/
    • Topic — Making a concept map is much harder than I anticipated, I went through a few iterations to get to mine. Also, I learnt quite a few tricks from looking at other people’s concept maps.
    • Thinking — It’s hard to present even the simplest ideas to beginners.
    • Useful — I found it very useful, to think about how I would present the topic if I had to teach it.
  • Jason Orendorff: http://teaching.software-carpentry.org/2014/01/21/concept-maps-software-best-practices-and-using-python-modules/
    • Topic — Did not learn a ton about my topics (outside of teaching them). I’m pretty familiar with them.
    • Thinking — Also not a lot.
    • Useful — I still don’t know if they’re terribly useful for me. The “best practices” map was for teaching and motivating, not for organizing a lesson, and that one might be usable. The one thing the “Python modules” map helped me do is find my implicit knowledge. Which is critical. But I would normally focus on motivating what I want to teach really really well, which I think would also order the lesson properly.

Understanding by Design (also called “reverse course design”)

  • start by writing the final exam
  • start with what you want them to know (the assessment)
  • then enumerate what skills they need to pass the exam. —> informs what and how you teach them.
  • Decide what success means first
  • akin to test driven design of software
  • can fail miserably if you “teach to the test” (think std’ized test, test-drilling)
  • determinants of whether learn:
    • are motivated?
    • do students feel as if they’re in control?

If you choose an answer to this question at random, what is the chance you will be correct?
a) 25%
b) 50%
c) 60%
d) 25%