I started working on a tool called "Promising Ideas tool," firstly named as "Big Ideas tool," one and a half years ago. It is an add-on to Knowledge Forum (KF), a knowledge-building environment used in many kinds of settings but mostly in primary schools. The Promising Ideas tool is really simple in terms of functionality. It looks like a lite and KF-confined version of Diigo or, which are both used for people to tag important information on the Web and leave traces for later-comers for better browsing heuristics.

However, the Promising Ideas tool was designed on a distinctive philosophy and to meet very different needs. A phenomenon I learned from both talking with teachers and reading student notes in KF is that, young students may repeat each other's ideas for weeks and weeks without moving their shared understanding forward. The reasons for this phenomenon are complicated. One possible reason might be that young students cannot intentionally process that much information produced by their peers at a time; as a result, although the community knowledge might grow fast at the beginning of their inquiry, it may enter a plateau phase when students are slowing picking up each other's ideas and promising ideas become swamped in a number of repeating ones. It seems beneficial for a class to stop contributing new notes at some point and start reflecting on existing ideas in their community. At this point, two questions came to our minds: (1) Do/Can young students have a sense about the status or frontier of their community knowledge? (2) How can we make promising ideas in their discourse more visible and help them make more progress?

So we designed and developed the Promising Ideas tool, to help students make promisingness judgments on their peer discourse. As soon as we produced the first prototype in 2010, we started to test the tool in a primary school. A paper[in pdf] about how well can Grade 5/6 students do promisingness judgments without any intervention was presented at CSCL 2011 conference in Hong Kong. From April to June this year, we conducted the most recent pilot in a Grade 3 class. Before we started this pilot, teachers were very concerned whether Grade 3s can grasp the meaning of "promising;" they were also worried whether promisingness judgments would lead to a fear of sharing ideas among students, as I described in a previous post. Those concerns are all valid and valuable and we agreed we should pay special attention to the words we use in communicating the tool and the philosophy and intention behind this tool to kids.

The first session of this intervention was a 30 mins Knowledge Building (KB) talk about the meaning of "promising ideas," led by a teacher who had years of experience in KB. This talk was fruitful. The class arrived at a shared understanding that promising ideas were ideas that "they wish to spend time on," "are not necessarily correct at the beginning and may end up change a little bit in further inquiry," and would "deepen their shared understanding." This promisingness talk is crucial for successful use of this tool, in the sense that it helps students understand the reason and purpose of using this tool. The first part of the following video, 0:01 to 4:40, is a concise version of the promisingness talk.


The second session of this pilot was a tool tutorial session led by my colleague Monica Resendes. She walked through all important features of this tool in front of all students. Questions popped up from time to time among students and were quickly responded. This session lasted for around 15 mins. Students learned this tool pretty fast and moved to do actual promisingness judgments with this tool. (To get a better sense about how this tool works, watch the following video.)


In the third session, students worked in pairs and spent 20 mins on conducting promisingness judgments on their notes in a KF view they were working on. They identified 88 "promising" ideas (28 distinctive ones after collapsing overlapping ideas). All those ideas were piped into an idea space attached to the specific view they were working on. After idea tagging was done, we displayed the idea space to all students sitting in a circle (see the second part of the first video). The teacher and students went through the ideas and discussed on each idea at the top of the list. Finally, they collectively selected three "most promising" ideas and exported them to a new view for further inquiry in the following weeks.

We ran this intervention for two iterations in the same class when they were studying the same curriculum unit. Eventually, students produced three KF views that represent three different stages of their collective inquiry. Preliminary analysis shows the quality of their notes (measured with a "scientificness" scheme) and students' conceptual understanding of that unit (measured by Latent Semantic Analysis between student notes and authoritative sources) has significantly improved across three views. I also analyzed data from a control group and did not find such significant improvements.

To summarize, the Promising Ideas tool is simple and can be used very differently. However, I think the most powerful way of using this tool is to facilitate promisingness judgments by students in their knowledge-building discourse. This use case or intention clearly distinguish the tool from some social bookmarking or highlighting tools, like Diigo and While the meaning of "promisingness" and fundamental psychological mechanisms of doing promisingness judgments are still open to debate, it is important to engage students in a discussion about the meaning and importance of "promising ideas" before actually using this tool. Substantial work needs to be done in the future!

Blog Logo

Bodong Chen



Crisscross Landscapes

Bodong Chen, University of Minnesota

Back to Home