 Thank you all so we're approaching actually the second segment and the way we thought how to transition from one segment to another is by having a transition talk so I'm very much looking forward to Justin Reich's presentation which is a transition presentation I think one of the topics coming out of the first segment or common themes I should say that we will revisit certainly in the second segment this afternoon and more on that just in a few minutes is how much context matters I think we already started with that in our early morning presentations the context of learning the context of education matters the context in which technology is used matters ultimately for the success and for the impact of open resources on education Justin Reich will focus on one particular aspect of where context matters and that's when open encounters different classrooms and we're really looking forward to it Justin Justin is a fellow at the Berkman Center among many other things he recently finished his dissertation at the Graduate School of Education here at Harvard and this leading fantastic efforts also working closely with teachers the EdTech initiative is among the project he's working on and thanks for being here and transitioning us sure thank you so I'm gonna start with a mutually beneficial shameless plug so I have a blog EdTech researcher which just got picked up by Education Week and so I'm gonna be publishing from that platform all of you build and research amazing things and when you do cool stuff you should let me know and then I'll write it up and then that's hopefully one way that people who are interested in communicating with education leaders can have one more platform to do that with so if you I just started on Monday so come and visit but also let me know what you're up to so that I can write about it and share it with other folks so my task is to serve as an inflection point between a morning of thinking systemically about the OER system and then an afternoon of thinking about specific case studies so when I think about thinking systemically I think about building a conceptual model a simplification of reality that offers us some kind of conceptual clarity and what I think case studies are really good for is introducing a kind of empirical messiness introducing particular details particular context that forces to go back to the conceptual models we build and rethink them in different ways and figure out what kinds of pieces we should add and subtract so I'm gonna tell you one story today about a case that I have been looking deeply into and how it sort of shaped and transformed the model of OER that I have in my own head as an example of the way that we might approach the work that we do together this afternoon so here's just a little schematic of the OER ecosystem as we've described it and challenged it this morning but there's a group of folks who consider themselves builders and suppliers who create resources and platforms and aggregators and other kinds of things for the most part they depend upon the internet as a distribution mechanism and they imagine on the other side of the series of tubes there being two populations of interests a group we've identified as facilitators in the group that we've identified as learners and the idea is that people create these OER whatever they are and they pass through the tubes of the internet and they arrive in the brains of young learners either directly or indirectly through some kind of facilitator and then what I think is really cool and really promising and exciting is that we can imagine because of the nature of our resources a kind of two-way interaction when when that OER arrives in the brains of young learners we can watch how they're using it and measure that we can wrap those things in peradata and we can pass that back through the internet and pass it back to the builders so that there's kind of an iteration that's able to happen in in development if you wanted a simple schematic of how the morning was laid out that's my kind of really basic schematic at the risk of running over my time I'll pause already to say that I think there've been some really great challenges to that schematic that have been brought up by some cases that we've looked at already so so things that I heard particularly from Vicki and her discussion of the flat classroom project and John and her discussion of the flip classroom model is this notion that we have three distinctive groups is really problematic the idea they're builders passing things to learners doesn't make a lot of sense unless we you know if we don't want to imagine facilitators and learners as consumers if we want to imagine them as co-constructors of their own learning experiences then maybe the whole supply and demand model is something that we could think about shifting and changing but that's not what I'll talk about because I just heard about this morning I'll talk about things that I've been digging into for the past four years oh let me say one more thing about this so there's a pretty simple model that underlies all this which is that as open education resources innovation happens the ideal result is at least in some linear way learning improves that the number of neurons and young brains that are reorganized for pro-social purposes is greater as innovation improves so that this is the kind of simplest possible schematic so the Hewlett foundation for the last four years has funded my research into wikis and how not mass media wikis but how classroom wikis are used in different kinds of settings and what that has made me very attentive to is the ways in which these this schematic might look differently if you imagine it working in profoundly inequitable school systems so one of the distinctive features of the United States and this isn't the case everywhere is that we have a profoundly inequitable school system so we talked a little bit yesterday about the idea that you know that the United States is kind of in the middle of the pack of OECD countries but if you disaggregate that data by wealthy and poor students what you find is that our wealthy students perform as well as as good schools all around the world and the results you know the out to the educational outcomes for students in low-income homes and families is shameful and embarrassing so for me one of the one of the signature features that's come out of my research is rethinking this schematic to take account profoundly inequitable school system so what I wanted to do is show you the kind of amendment that I've made to this kind of general schematic and talk about how I think wikis sort of illustrate the importance of thinking about the importance of thinking about how all of the stuff all the work we do operates in profoundly inequitable school systems so I'm going to make one really simple conceptual move which is that I still imagine that there are a whole group of builders I still imagine that there's a internet which is the kind of distribution mechanism but on the far side of the internet I think it's really useful in a simple way just to imagine that you have two parallel systems one serving low-income facilitators and learners and one serving really affluent facilitators and learners so there's the nine million dollar science building that was just built at the Deerfield Academy which is my alma mater but the hope is that OER is built by builders and it passes through a neoliberal democratizing internet and it travels into both of these environments relatively equitably in people benefit and we could actually tell two kinds of stories about the results of that OER passing to these different kinds of environments so one kind of story that we could hypothesize is that in the beginning we expect there to be disparities and learning outcomes for affluent and low-income students because we live in a profoundly inequitable society but affluent students already have access to all kinds of proprietary tools resources platforms aggregators all kinds of stuff and so we we build a bunch of openly accessible and freely accessible stuff it disproportionately benefits low-income students because they now have access to a whole wide variety of things that there are more affluent peers about access to for a long time we might call that the closing gap scenario another scenario you might imagine is that again we start with sort of a disparity based on society but in fact for a wide variety of reasons students in affluent homes and schools have a greater degree of social capital human capital technical capital and those kinds of institutions that allow them to take up and take advantage of OER resources more easily more profitably with greater efficacy that can happen in schools and learning environments serving low-income students and so actually what happens is that the widespread availability of freely available resources ends up disproportionately benefiting those who are already affluent that when you take a bunch of stuff and you just kind of put that stuff out there institutions with greater capacity take it up faster and so the technology serves as an accelerant of inequities that already exist in our society so we could call that the rising tide scenario from the you know the sort of famous Rawlsian phrase a rising tide lifts all boats that I'm not saying that anyone is harmed in this scenario but a rising tide lifts all boats but the mega yachts get lifted higher than the dinghies do so it was this case study of wikis that helped me think about this and frame this and let me tell you a little bit of the data that I found about how wikis are used in different settings so I study wikis that are produced on a platform called PB works so they're not it's not wikipedia it's not mass media wikis it's wikis that teachers create for a particular purpose or students create or small groups of learners create one thing that's awesome about studying cases in our field is that the range of data available to study any given case is you know mind alteringly fascinating and exciting so I had a chance to draw from a population over a quarter of a million publicly viewable education related wikis that have been created on PB works and for every you know and you all know this for every one of those wikis I had the entire edit history that we were able to examine where we could look at every interaction between every teacher and student in those kinds of environments and this is all just recently published in educational researcher in a paper called the state of wiki usage in US K-12 schools and one of the central questions that we asked was does wiki usage differ in schools serving different populations so one of the things that we did was we developed an instrument to measure the degree to which wikis provide opportunities for students to develop deeper learning competencies like expert thinking complex communication and new media literacy and we applied that instrument to a sample of wikis that was randomly drawn from this whole population so we have hundreds of thousands of wikis we draw a random sample and we identify in particular wikis that are used in US K-12 public schools so we can measure the kinds of learning opportunities that students have on these wikis and we can correlate those learning opportunities and that quality measure with socioeconomic status data from the public schools which is well kept in the United States from the common core of data so in a short talk I'm giving you a pretty brief summary of this but let me give you sort of one slide of data about this so one way to slice the sample of wikis that we looked at is to put them in four categories one category is failed or trial wiki so these would be wikis that like say hello world across the top and they're never changed again or welcome to Mr. Reich's world history class and that's the last edit you see another kind of wiki would be teacher-centered content delivery wiki so this is instead of handing out my syllabus to you at the beginning of the year I put my syllabus on a wiki and I put some links and things like that but there's no student interaction in the learning environment it's just a one-way content delivery device a third category of wikis would be individual student-owned wiki so you can imagine that if you assigned me to write a paper about Hamlet instead of handing you that paper on a piece of paper I post it on a wiki page and put on some links and put on some images and so forth or maybe you asked me to maintain a portfolio of my work across a year or multiple years and then the last category of wikis is kind of what we imagine wikis to be which would be collaborative student known wikis so this would be topical encyclopedias or AP history reviews or collaboratively produced choose your own adventures so one interesting question to ask is how does the distribution of wikis in these four categories differ between low-income schools and mid to high-income schools and for those of you are familiar I'm using the title one status is the distinguished the distinguishing feature here so what you find the most relevant piece for this talk is that wikis created in low-income schools are significantly more likely to end up in the failed or trialed category then wikis that are created in wealthy schools and wikis created in mid to high-income schools are much more likely to fall into this individual student own wiki category where we actually have students who are doing the creation doing the you know the the development of new media skills the development of performances of understanding it's sort of a tune for another time that actually only about one or two percent of wikis end up being used actually is what we imagine wikis being as collaborative learning environments but it but it was it you know this is a brief view but a powerful to me that essentially wikis are significantly more likely to provide opportunities for deeper learning if they're created in schools serving mid to high-income students another thing that we measured along the way was sort of how long these wikis persisted so what was the length of time between the very first change and the very last change the median lifetime of wikis created in low-income schools about six days the median lifetime of wikis created in mid to high-income schools about 33 days so not only do students have more opportunities with wikis that are created in mid to high-income schools but they persist longer and so you know presumably those those opportunities persist longer as well so if we go back to our schematic you know what what one way to one way to visualize this is that we have PB works wikis that are created by builders and that's you know this or the platform is created by a builder passes through the internet and makes its way relatively easily into high-income schools but for a wide variety of reasons of technological social pedagogical reasons the the platform stumbles in making its way into high-income schools so in in schools serving low-income families or low-income neighborhoods there's more test pressure teachers have more students per teacher they have fewer prep periods they have more classes they teach they have fewer curriculum supports available to them at a school level in a district level they have weaker technological infrastructure so for a whole wide variety of reasons the same thing which is created you know and then distributed using what you might call a dump and hope strategy I mean PB works is essential strategies we're going to build something we're going to put on the internet and hope that somebody picks it up get picked up really differently in in different kinds of places that I'm only able to summarize and then actually sort of the terror that I have about the scenario is that PB works actually has no idea that this distribution is happening but as they start collecting usage data about how thing out how PB works is is being used that pair of data is really being generated primarily by wealthy settings and so what you potentially develop is this really kind of terrifying feedback loop where we produce stuff we pass it through the internet we're not exactly sure who's picking it up but it turns out that affluent students are primarily picking it up they're the ones who shaping the usage data that we're collecting that usage data gets passed back to builders and we create this feedback loop where builders unknowingly are essentially designing for students in already advantaged locations that is that's kind of my terror of education technology and and so forth good I think I'm going to end right on time so in the case of wikis we you know if you imagine these two stories it seems pretty obvious to me that wikis you know sort of nurture a kind of rising tide sort of scenario where wikis disproportionately benefit the already affluent I'm not saying that that's the case with all different kinds of OER resources and education technology interventions although it seems to me that interventions that use distribution mechanisms similar to the ones of PB works might very well have similar kinds of outcomes like if the if the idea is we put this stuff out there and hope that people pick it up that institutions with greater technological social and human capital may be the ones who are more likely to pick it up so if you're at the Berkman Center you can't just stop with the empirical data you've got to think a little bit about what the normative normative and policy implications are so if you think this is true and first of all people should do lots of other case studies and we should listen to the case studies this afternoon to see if it's true what would you do about it well one thing you could take is a sort of neutral approach if there's more net learning who's ever learning that's a good thing it's not our responsibility that we live in an equitable society let's build the best learning objects we can and let's pass them out there another response you could have is we should actually at this sort of vulnerable moment where we're trying to promote open policy we should prioritize serving middle and high income learners and develop the kind of political support among those constituencies to get them behind the kind of work that we're doing you might not be surprised to find that my sympathy is lied with a third answer which is that education is the civil rights issue of our time and we should rethink distribution and delivery mechanisms to ensure that innovations target those who need the most support I've had a great time talking to Mike from road trip nation and how much they've been emphasizing delivering their curriculum to people in places like juvenile correction institutions thinking about their distribution mechanisms is not just getting it out there to anyone beginning developing something and distribute it to those who need the most so that's my story about how looking at a particular case study reshapes my thinking of the OER ecosystem I personally hope that one thing that you take away from it is this idea that it may be really useful to imagine profoundly inequitable school systems when we're developing our kind of overarching models of OER but in terms of thinking about the rest of the afternoon I hope that what we can all do is sort of take whatever model that we've been developing over the course of the morning whatever conceptual framework we have look at the particular case studies that will look at for the rest of the afternoon and see how they challenge how they influence how they command us to add or subtract different elements from those models so thanks a ton for your attention and hopefully that's a good provocation for the afternoon