Project OpenComps

This semester I'll be taking my comprehensive exams, or "comps." As a first-generation college student from the working-class rural Midwest, this is pretty unknown territory for me. I remember being a naive undergraduate who had to ask what masters and doctorates were, and when I started my PhD program I had to ask similarly naive questions about the mysterious and vaguely threatening-sounding comps. Quite simply, comps is my opportunity to show a committee of faculty members that I have the knowledge and skills to take on my own research -- namely, my dissertation. Yes, there are written and oral examinations, but it's the process of working with a committee of faculty to both narrow my focus and double-check that I know what I should know that makes the process valuable.

Thankfully, I've been able to watch other graduate students prepare for and take their comps (usually passing, but not always) and now it's time to prepare for mine. I'm going to share that process and preparation with you and tag things #opencomps along the way. You might consider this a step in the direction of something like Hack the Dissertation. I've learned that the entire comps process can vary from program to program, so I can only really describe what it's like for a math education student in CU-Boulder's School of Education. Let's recap how I got this far:

• With a BA in Mathematics (Teaching) from the University of Northern Iowa and six years teaching high school math, I decided to go to grad school. Having missed the admissions deadline for a master's program, I spent a fall semester as a continuing education student and was admitted into CU-Boulder's master's program for the spring. It went remarkably smoothly, thanks to the help of my advisor David Webb.
• I expressed an interest in the PhD program and was encouraged to apply. I got recommendations from my current professors and good (enough) GRE scores to be accepted. This meant abandoning the master's program, but thankfully many of the credits I earned transferred to the PhD program.
• My first year in the PhD program was spent in the "core," the set of six courses every cohort of incoming doctoral students take in the School of Ed. Those courses include two semesters of quantitative methods, two semesters of qualitative methods, a course on theoretical perspectives on social science research, and a course on education research and policy. I took a seventh core course, covering multicultural education, the first semester of my second year.
• I focused the rest of my second-year coursework on two areas: math education (a course on algebra and a course on theories of mathematical learning) and educational measurement (a course on survey research with an introduction to item response theory, and an advanced measurement course with more IRT and generalizability theory).

I'm required to have 56 hours of coursework (not including dissertation credits) for a PhD. In some programs you need to finish those classes before comps, but in my area it's okay to just be close to 56 so long as the coursework provides the necessary foundation. With over 50 credits under my belt my advisor says I'm ready, so I've taken the first two steps this semester towards comps. First, I needed to choose a committee of three faculty members. My first choice was easy -- my advisor David Webb. I can trust David to make sure I'm ready in the areas of math education and classroom assessment. Also on my committee is Derek Briggs, who will surely hold me to task in the area of quantitative methods, validity, and causal inference. Derek didn't actually teach my core quantitative classes, but I took my measurement courses from him and enjoyed working with him. Due to my wandering interests, the third choice wasn't so easy. (Someone in policy? Qualitative methods? Stats ed? Learning sciences?) I went a bit onto a limb and chose someone I've never taken a class from: Bill Penuel. I got to know Bill a bit last spring during some facilities work, and I'm working for him this semester on a project that combines many of my interests: math ed, professional development, technology, pedagogy, task design, and assessment. I like what I've seen of the project so far and think working more closely with Bill will be a very good thing.

The second step I've taken towards comps this semester was to assemble a reading list. Basically, the reading list contains what I've read for my classes and what I've cited in papers and it gives my committee a place to look for holes in my knowledge. Thanks to Mendeley and careful curation over the past two years, the list wasn't too difficult to assemble. It's long and looking at the 40+ pages of references made me not feel so bad about not reading much over the summer. Take a look at my reading list for yourself, and feel free to ask about anything there, or suggest something you think might interest me!

Nielsen's Reinventing Discovery (2011) in the Context of Education Research

As a Ph.D. student I've taken my share of methods courses, giving me skills in everything from ethnography to ANOVA. But as important as those things are, I've sensed that there are new research methods emerging thanks to technological advancements and online communities. Our lives are too data-rich and our means of communication are too plentiful to limit ourselves to the same methods for research -- and learning -- that we used just 10 years ago.

Even though I feel like I live in the thick of this revolution, engaging with teachers and researchers on Google+ and Twiter, I wanted a broader perspective on how researchers use networks to make new discoveries. For this I turned to Michael Nielsen's book Reinventing Discovery: The New Era of Networked Science. Although Nielsen is a pioneer in quantum computing, I hoped to find some ideas that I could apply to a social science like education research.

Nielsen uses a variety of examples and concepts to describe what works and what doesn't (or hasn't) in networked science. Instead of listing them here, watch this TEDx talk by Nielsen:


If that talk wasn't long enough for you, Neilsen held a longer talk at Google that is worth checking out.

As much as I like Neilsen's example of Tim Gowers's Polymath Project, I can't imagine a direct translation to education research. One of the beautiful aspects of mathematics is that it usually doesn't require conducting an experiment, interviewing subjects, sampling a population, or agreeing on a conceptual framework -- the kinds of things that make social science untidy and difficult. Frankly, if solving problems in education were structured like proving mathematical theorems, I think we'd be solving more problems and finding better solutions than we are currently.

Neilsen's story about Qwiki hits home for me. For some time now I've imagined creating and maintaining a wiki that essentially translates the contents of the NCTM's Second Handbook of Research on Mathematics Teaching and Learning into knowledge that teachers could access and use. Just like Qwiki, it's easy to get math teachers and educators to agree that this would be a great resource to have. Unfortunately, I'm not sure how a math education wiki like the one I've imagined would avoid Qwiki's fate. Without incentives for experts to contribute and maintain the site, I'd probably spend more time fighting spam than helping teachers.

Neither the Polymath Project or Qwiki offer a blueprint for a new kind of mathematics education research. Thankfully, Nielsen describes some general characteristics for successful networked science. First, in his chapter titled "Restructuring Expert Attention," Nielsen suggests networked science has these attributes:

• Harnessing Latent Microexpertise -- The project must allow even the narrowest of expertise. A 3rd-year algebra teacher might not have the broad expertise of an experienced math education researcher, but that 3rd year teacher might have small elements of expertise that exceed that of the recognized experts.
• Designed Serendipity -- The project needs to be easy to follow and encourage participation from a variety of experts. You want problems to be seen by many in the hopes that just a few will think they have a solution they wish to contribute.
• Conversation Critical Mass -- One person's ideas need to be seen by others so they create more ideas, and the conversation around all the contributions keeps the project going.
• Amplifying Collective Intelligence -- The project should showcase the fact that collectively we are smarter than any one individual.

Those are all great characteristics of any project. But what makes this any different than any traditional, offline project? Nielsen offers several suggestions. Unlike a large group project with clear divisions of labor, technology allows us to divide labor dynamically. Wikipedia certainly would not have grown the way it did if labor had been divided statically between a set of contributors. Also, networked science uses market forces to direct the most attention to the problems of greatest interest. Lastly, contributing to an online project rarely feels like committee work, and participants can more easily ignore poor contributions or disruptive members.

Projects like Wikipedia and Linux exhibit the above attributes, but Nielsen explains that such projects needed something extra in order to scale to thousands of participants. Nielsen describes these in a chapter called "Patterns of Online Collaboration," and they are: (1) being modular, (2) encouraging small contributions, (3) easy reuse of earlier work, and (4) signaling to what needs attention. When I look at this list and think of Wikipedia, I can see how well a wiki or open source software project fosters these patterns. But how do we build such a project in education? Given Nielsen's framework above, a project that would interest me needs three key aspects:

• The content of the project has to be something that both teachers and researchers can contribute, such as a collection of math tasks, curriculum plans, or perhaps pedagogical techniques.
• Teachers need to be able to easily use and modify each other's content.
• (This one's the crux!) When teachers use content, there needs to be a way to collect and submit feedback about the use of that content, and that feedback becomes data that researchers can use not only to improve the content of the site, but to produce new and traditional reports of research.

It's that last bullet that's the hardest but most intriguing. There are so many places to get lesson ideas on the internet, but I don't know of any that collect data about the effectiveness of the lesson in a format suitable for research. Khan Academy claims to do this this kind of data collection internally, but KA is a closed project that lacks nearly all of the attributes Nielsen has described in his book. The project I want needs to be an open one, with all of its moving parts exposed and no more owned or identified with a single participant as Jimmy Wales is identified with Wikipedia. If you have ideas for what such a project could/should look like, leave them in the comments!

Peeping Tom: Finding Windows in the Ivory Tower

Today on Twitter, Tom Whitby posted:

I understand the intent of Tom's post: we have too many teachers who have become detached from some of the "big thinkers" in education. It's easy for a teacher, with all the pressures and responsibilities, to become isolated in their classroom with their students. Fortunately, it's easier than ever to traverse the branches of the internet and find leaders in education online, as well as other teachers who want to share, discuss, and debate big ideas in education.

While I'm sure Tom didn't intend for his list to be all-inclusive, all the names listed have something in common: none of those people are current professors of education. I'm not saying that professors have cornered the market of good ideas, but rarely do I see them mentioned on Twitter or elsewhere outside the ivory towers of academia. (Not suprisingly, several professors who are breaking down this wall are professors of educational technology, such as Alec Couros and Scott McLeod.) Trust me: ed school professors care just as deeply about students, schools, and the improvement of our educational system as anyone, and many have wonderfully big thoughts and ideas. In addition, they have a scholarly duty to promote ideas that have been tested and shown to have positive effects, not just ideas that sound like good ideas.

This might not be the place for a lame sports analogy, but I'm thinking of it this way. I love baseball, and I could happily spend hours listening to Bob Costas and Peter Gammons describe the nuances of the game. But if my job is to walk up to the plate and hit a major league fastball, do I want Costas or Gammons as my hitting coach? No! Give me Joe Varva or Rudy Jaramillo. Never heard of them, you say? Well, they're both major league hitting coaches, for the Twins and Cubs, respectively. Costas or Gammons could probably help me get a swing that looks like Joe Mauer's, but I'd need Varva or Jaramillo to help me develop my best swing, not one modeled after somebody else's. And neither Varva or Jaramillo themselves played in the majors. They know what they're doing because they tirelessly treat their jobs as a science. Alfie Kohn isn't Joe Varva. He's Peter Gammons -- an intelligent and thoughtful commentator who is making positive contributions to his profession and our enjoyment, but not necessarily a scientist.

So while you're asking your colleagues about Kohn, Marzano, and Sir Ken, try asking them if they have heard of Linda Darling-Hammond, Deborah Ball, Michael Apple, Truus Dekker, Alan Schoenfeld, or Lorrie Shepard. Don't know them? You should, but if you don't, don't be too hard on yourself. I was disappointed to the see that the list of Race to the Top scorers was heavily populated with educational consultants, institute founders, foundation advocates, and others who might profit from the results, instead of more ed school researchers. So maybe Arne Duncan doesn't know many of the names on my list, either. But it's not all his fault, and not all your fault, either. Our system of higher education and scholarly publishing is holding up those ivory walls, walls that work both ways. Stick to Alfie Kohn and let the wall crumble, or read Linda Darling-Hammond and try to knock it down.

Managing the Flow of Information in Your PLN, and Why You Should Stop Buzzing Your Tweets

While I've blogged since 2001, I resisted social networking tools for a long time, probably the result of seeing my students use MySpace. To me it looked like Geocities, only more judgmental and vain. I finally joined Twitter in the fall of 2008, Facebook in spring of 2009, and last fall I took steps to separate personal and professional networks. Now I use all kinds of social services, and often find myself thinking more about how to use them than actually using them. Hopefully in this blog post all that thinking will pay off.

Just like there's no one way to establish friendships, there's no one way to build a PLN (professional learning network). Here are the primary tools I use, ranked in order of importance:

1. Twitter (there's no easier way to connect and converse with others in real-time)
2. Blog (everyone needs a flexible place to express original thoughts and receive feedback)
3. RSS Reader (for finding and keeping up with other people's content, although some just use Twitter for this purpose)
4. Social Bookmarking Service (for saving the interesting sites/pages you find)

Like I said, those are my primary tools. There are many more out there, and new ones show up all the time. As we add these new tools to our toolboxes we gain powerful new ways of connecting with our peers. Unfortunately, that power is diluted when the interaction between tools gets complex, as it has with Google Buzz. A number people in my PLN have tried to integrate Buzz into their PLN with mixed success. Buzz has too much promise to ignore, so let's learn from our experience and think about how we're using Buzz.

People will remember the privacy mistakes Google made when they launched Buzz, but I think they made another mistake that has gone unnoticed: instead of importing tweets into Buzz, Buzz should be using Twitter to notify people of new buzzes! Buzz is bigger and more capable than Twitter, and I think importing tweets will prove to be a backwards flow of information. If your blog had the option to import each of your tweets as a new blog post, would you do it? Of course not. You wouldn't use Diigo to bookmark your tweets to share them, either. Twitter asks you, "What's happening?" and while we've used it to do much more, tweets aren't proving to be at all useful in Buzz. We've all used Twitter to notify our PLN about our new blog posts; let's use Twitter the same way to alert our PLN about a new buzz.

Here's my attempt to diagram the flow of content in my PLN:

The model is far from perfect, for sure. You can save and share in both Diigo and Google Reader, but the services are different enough that I use each independently for each interesting site I want to share. (Thus the reason for the dotted line.) The "Share to Reader" bookmarklet gets pages into Buzz, and I would personally be happy if Google bought Diigo and really worked to nicely integrate its capabilities into Buzz and Reader. Also, Google really should integrate blog post comments in Buzz to Blogger, and vice versa. We really need a unified commenting system in a lot of places, but there's no excuse for Google to not have this figured out for their own services. Lastly, Twitterfeed shouldn't be necessary, as I think Buzz should have the option to post to Twitter.

So, in summary, remember that 1) there's more than one way to build a PLN; 2) tweet a buzz, but don't buzz a tweet; and 3) expect some bumps along the way. (And try not to stress out about the tools as much as I do!)