In need of a Base Case
Comments about coding, research and life

Last week Stella Vosniadou from the University of Athens visited campus to give two talks as well as prepare for being here next year as a visiting faculty member. This post contains some reflections and ramblings from the notes I took from her talk.

Stella’s work on conceptual change is built on the foundational idea that by the time kids start primary school they have naive constructions of knowledge. These constructions are built from students’ interactions with everyday life and can be found echoed from their home environment and culture. The models that children and novices build of the world are based in ontological and epistemological presuppositions.

The naive models that students/children hold can be described as observation and cultural information about the phenomena, beliefs held by the child, and mental models of the interactions of the artifact with the world.

Stella works within a framework theory[1] which says that knowledge is built into frameworks or models, and that significant conceptual change occurs when the framework itself is restructured. Stella has done a lot of work with children’s understanding of the day/night cycle and also of the seasons. She has worked to categorize the types of misconceptions that students have about day/night based on several different models of the earth and sun.

When we run into trouble as educators is when we assume that by simply presenting students with new information that we can break apart a strongly held framework and restructure their model of interaction. Often students see things learned in school as inert knowledge - information that does not have any application to everyday life (when am I going to use this?)

Recently, in a study of software engineering misconceptions of students here at CMU I encountered situations where students believed both a misconception and its inverse to be true.
This implies to me that they are in the fragmentation stage of conceptual change - they are beginning to see that restructuring their framework is necessary and are either constructing a secondary framework, which will overtake the first (incorrect) framework, or replace it completely. (We are not sure if the original framework is restructured to the new, or if a completely new one is built and then becomes the dominant structure called upon when reasoning.)

How does this fit into my other work? I am looking at naive understanding/mental models of data and how that impacts students’ ability to reason about and write code for problems. I believe that our students have particular naive models and those models are the framework which they use to reason in our introductory courses until we provide them with new reasoning points (think bottle cap arrays or analogies[2]).

More to come…

[1] Vosniadou, S., “Capturing and Modelling the Process of Conceptual Change”, Learning and Instruction, Vol. 4, pp. 45-69, 1994

[2] Meyer, R., “Models for Understanding”, Review of Educational Research, Vol. 59, No. 1, pp. 43-64, 1989

The Obama administration released the National Educational Technology plan this week. You can find the entire document at http://www.ed.gov/technology/netp-2010. Below is a quote from the executive summary that just highlights the importance of creating strong technology and computer science standards for our country.

“How we need to learn includes using the technology that professionals in various disciplines use. Professionals routinely use the web and tools such as wikis, blogs, and digital content for the research, collaboration, and communication demanded in their jobs. They gather data and analyze it using inquiry and visualization tools. They use graphical and 3D modeling tools for design. For students, using these real-world tools creates learning opportunities that allow them to grapple with real-world problems – opportunities that prepare them to be more productive members of a globally competitive workforce.”

There are some good beginnings to computational thinking here - the use of modeling tools (btw - 3D modeling tools - that would be great for the Alice team..) and inquiry and visualization.

Now to put my head back down into the research work. Sorry for the lapse in posting, but trying to move forward. I’ll try to do better.

Data from the NEAP HS Transcript Study is reporting that students are taking more STEM courses in their completion of a high school diploma. Yet we continue to struggle with enrollment numbers in computer science. Maybe thats the wrong sentence. Perhaps it is that we continue to struggle against the problem of starting and maintaining successful HS computer science programs.

I dont think anyone would argue that there are wonderful programs out there, mostly because of the FANTASTIC teachers who run them. But we have yet to figure out how to bottle that to grow that model into new schools, or schools where the program is waning.

Within the data from the NEAP report you can see that the computer-related course taking has actually increased. I’m not sure how they judged computer-related however. There are two coding areas from the CSSC course codes that were used that apply to computer science - Both 07-03 (07 is the Business heading and 03 is the particular part which includes CS) and the entire section of 11 (although this also includes courses like “computer appreciation”, “computer math”, Introduction to the Internet”, as well as APCS.

Unfortunately neither of those subjects are available using the online data tools. I should make a request to get at the data and look at what happens when you remove the “applications” course codes from the data and re-calculate the graph on the bottom of page 9 of the publication.

More about this, I promise..

Yes, I know its been a while, and will probably be at least another week until I pick up regular postings again But, I wanted to share this quote from my Intro to CS Education course that I taught this semester:

This course raised the bar for computer science education, both in expectations on the students and in the teaching of the course itself. Before entering this class, I thought education was synonymous with communication, and that to be a good educator it sufficed to refine clarity in communication. After taking this course, I am amazed at my naivete. I now have an appreciation for assessment, metacognition, motivation, engagement, cooperative learning, cognitive architecture and many other concepts that contribute to good teaching.

This was written in one of the online course evaluations that are done at the university level and I think it sums up a resolution of my goals for the course.

I’ve always been impressed by the tradition of applauding for the pilot when you land in Europe (at least I’ve experienced this in Greece and Italy when I traveled a few years ago). On this trip, there was no applause upon landing and it prompted some thoughts that have made their way here (in front of all of the readings that I need to write about).

Why do we applaud things? We applaud them because they are unusual, because they are exceptional, or in appreciation. As things become common place we are less likely to applaud what would have caused us to applaud early on. Have we stopped applauding computer science?

New developments, new applications, new inventions - while applauded within our own community do not often make the larger news in current times. I wonder what affects that has?

Just some quick notes - more about Koli perhaps on the way home.

Starting with Mark Guzdial’s blog on What Changes CS Education? to Alfred Thompson’s post on What does it take to make a change in CS education? there has been an interesting thread of conversation going on. Because the semester is in full swing and I have a LARGE machine learning homework out right now I originally started just a comment on Alfred’s blog. Well.. For any of you who know me, I am very passionate about this subject and after writing a few paragraphs figured it was getting too long for just a comment.

So to start, here is my comment on Alfred’s blog, with more, new comments below:

::sigh::

Forgive me for being an education purist, but I think this is part of the reason why we are in the pickle that we are in as a discipline. Relying on tools or the industry to define the introductory experience, and even more so the novice experience is .. well.. crazy? ineffective? ridiculous? not sure what the right word is.

When the biology department tries to decide what biology knowledge makes someone a literate citizen they don’t look at what professional biologists are doing. They don’t look at what nifty tools are available for teaching biology. They decide what it should mean for your average American to have an understanding of biology. They define a set of broad standards that would allow a knowledgeable teacher to implement those standards in a classroom in one of many different ways, potentially using one or more of the tools available to them. Then they train the teachers. To BOTH have an understanding of biology AND an understanding of what it means to teach biology - both to be familiar with the standard curriculum and the pedagogical nature of the course.

How is this different from what we do? We argue about what the right “foundation” is. College CS people argue about why AP is not good, but the AP program is one of the few programs with a solid curriculum, that does TEACHER TRAINING. While there are great things being done in a few places, its often hard to find something (outside of AP workshops) that a new/novice teacher can hold on to, and take back to their administrator as “standard”.

We need to stop arguing about language, tools, etc. and decide what it means for the AVERAGE American to be literate in computing. (notice the computing - not the computer science, its a bigger question than just CS) Then we need to write a set of standards, or agree that the ACM’s standards are good. There needs to be consensus in the message coming out from our community. Then there are the economics to consider. How are schools going to pay for this?

and now for the new part

It really does boil down to economics in the long run. We can train 10,000 new teachers, and ACM can publish excellent standards that are used by every CS teacher in schools, but unless we somehow make it cost beneficial or cost neutral to schools - those 10,000 teachers (for the most part) will not find a job unless they are filling a retiring teachers spot. There are two types of economics going on here. First, the money issue, how in the world are schools going to afford to pay for a new teacher for a course that is not required, and currently doesn’t exist in the school so there is no demand for it?

Secondly, the economics of the student’s schedule. Students’ days are currently full. Any high school teacher can tell you that. The days we remember of study halls in high school or “free periods” are a thing of the past. In the competitive college market students have taken to PACKING their schedule with as much as the possibly can, with the largest percentage of AP classes they can to try and get into the best schools with the best scholarships. As of right now, college admissions officers give no more weight to AP CS than any other ELECTIVE AP. So why take CS, when you can take a language, art, music or one of any other AP courses? And if it is not an AP course why take it when an AP course has more weight on your transcript?

So, there is no money to hire teachers for this elective course. And there is no student demand for the course. How does picking a better tool fix the economic problem we are facing? How does writing a better textbook fix this? Yes, it may change what is happening in colleges, or in a few established HS programs - but there are bigger issues here.

I have some ideas about this, but to save time I’m going to leave this with the questions. (Machine Learning - homework calls) But as we consider whats going to make the LARGE changes, in the location that we all agree is where it needs to be (K12), the economics are undeniable and unavoidable. Lets keep those in mind too.

This morning I read an article about Sarah Fine and the reaction to her decision to leave an inner city DC school. No one questions that Sarah is an exemplary teacher, and her decision prompted both outrage that she should “abandon” the class as well as understanding.

It reminds me a little of some of my own challenges - when I first decided I wanted to teach, that it was important to me I had to justify this decision with several people - including my then significant other who thought that was a waste and that I should go into the computer industry. The number of times I heard the phrase “those who cant, teach” made me question my decision to the point where I actually interviewed with several companies for systems administrator positions.

I started teaching in an inner city school in Ft. Lauderdale FL. 2000+ students and I was a minority. I had a student threaten to turn his gang on me for trying to enforce school policy. (that was on the second day) I did eventually leave that place due to the communities reaction to a tragedy that occurred. I did keep teaching, but I moved to a suburban school district with different problems.

I was burned out after two years, and if I hadnt had the option of moving to a better district I’m not sure how much more I would have lasted through. Now, as I hope to make a change, to contribute to a community of educators in a different way I again face questions. Only this time its along the lines of “why are you leaving? your students need you”.

It feels like a double standard. First teaching isnt good enough for intelligent, hard working people (clearly they should do something with better options for success) and yet, when they can take no more of the lack of support and difficulties that Sarah describes they are berated for leaving. We need to let our teachers know we value them - right from the time they decide to be a teacher. Maybe then more smart, excited people will pursue and stay with teaching.

If you have a teacher in your life (either a colleague or one of your childrens’ teachers) who is doing a good job, let them know you are glad they made the decision long ago to be a teacher. One comment like that used to get me through a whole semester.

One of the things I love about being at CMU is the ability to see talks given by the professors here (and even sit in on some of their classes). One of the favorite lecturers in the CS department is Luis von Ahn. Luis developed an entirely new field of computer science called “human computation”. Here is a great video of Luis at the CCC event at the library of congress talking about how his research has changed the world.

I think this is a great thing to show students as it really starts to give them a window into the idea that computer science is not all about the programming - yes there is programming involved in a lot - but it is so much more than that.

I know that there are a large number of people out there doing excellent work to encourage girls in computer science. A new US Department of Education program named Women’s Educational Equity. This has prompted 13 groups to apply and be awarded this funding. An interesting article here - is one of these institutions near you? Any of them doing computer science or just more general STEM?

I have been involved in the creation of the Sci-Tech high school in Pittsburgh. I’m not 100% positive but I believe some of the grant money awarded to Pgh public schools is going to the sci-tech program. One of the “strands” through the high school is computer science, and its interesting to see how this is growing.

Wow. Just go and read the press release. This is very exciting and hopeuflly a sign that more and more people are recognizing that K12 has a very large place in the discussion of computer science education. And on the other hand, it gives me hope that the STEM supporters are recognizing that something other than technology usage (like word and powerpoint) need to make up the T in STEM.