Thursday, May 14, 2009

Jeff Howe on Crowdsourcing

There are numerous insights in Jeff Howe's book on crowdsourcing about how and why to engage people in building websites.

"Given the right set of conditions, the crowd will almost always outperform any number of employees-a fact that companies are becoming aware of and are increasingly attempting to exploit" (p. 11).

"Croudsourcing has the capacity to form the perfect meritocracy. Gone are pedigree, race, gender, age and qualification. What remains is the quality of the work itself" (13).

Howe relates a story by Linda Parker about the Cincinnati Enquirer in regards to their efforts to get the crowd involved in contributing stories. She said, "It used to read, 'Be a Citizen Journalist,' and no one ever clicked on it. Then we said, 'Tell us your story,' and still nothing. For some reason, 'Get Published' were the magic words." Howe notes, "There's a valuable lesson here: people want a voice, but that doesn't mean they'll use the vernacular of journalism" (106).

Howe, J. (2008). Crowdsourcing. New York: Crown Business.

Wednesday, May 13, 2009

Applying Crowdsourcing to Family Life Education

Based on Jeff Howe's book on Crowdsourcing, here are some brief applications of his ideas to family life education.

1. Pick the right model of crowdsourcing. Crowds can help provide wisdom about family life, they can help create the content or educational process of family life education, they can rate family life educational material on matters of relevance, importance of helpfulness, they can contribute financially to the work or some combination of all of these strategies. (More about models of crowdsourcing applied to family life education.)

2. Pick the right crowd. Howe suggests that a good crowdsourced site has 5,000 active participants, but they need to be engaged in your work. Designers of family life education websites needs to design for specific audiences. Too often family life education web designs don't have a specific audience.

3. Offer the right incentives. Howe writes, "With few exceptions, the most important component to a successful crowdsourcing effort is a vibrant, committed community" (p. 282). Fostering and sustaining a community of interested partners means creating a process that rewards the community of users. What types of rewards do parents or family members want to receive for participating? A chance to help, a chance to share with others, what? (Other notes of community building. )

4. Crowdsourcing is not cheap or easy. One of the myths of crowdsourcing is that the web designers have less work to do or it takes less money to foster community-based web development. Not so says many who have done this work.

5. Crowdsourcing is a partnership between good management and the participants. Crowds don't self organize and manage. Good crowdsouring models are most effective when they have good leadership. That is, when there is a model that provides easy direction and opportunity for contributing.

6. Find the right level of contribution. Howe writes, "any task worth doing is worth dividing up into its smallest components" (285). Effective crowdsourcing is finding an appropriate-sized unit of contribution that is manageable for someone to do and provides a building block for the overall project. (See my comments on microlearning. units.)

7. The crowd is likely to produce a lot of junk. It is naive to think that crowds will only create wisdom and great products or that all members of the crowd will have the same talents. Effective crowdsource development means having a way of finding the best material and fostering the best talent for the specific jobs you need. (See more about the impasse of Sturgeon's Law.)

8. The crowd will also contribute some value. There are people who will help you build more effective and useful family life education websites.

9. If you are lucky enough to develop a strategy that involves a crowd... listen to them. Trust their guidance.

10. "Ask not what the crowd can do you for, but what you can do for the crowd" (p. 287). I am sure that Howe put this last so that the potential website developer would be left with this one thought. For crowdsourcing to work you have to find a project that you think is valuable and that the crowd thinks is valuable. The more the project serves the needs of the crowd and provides them with engaging, interesting, rewarding, and meaningful opportunities the better chance you have of success. Family life education is ideally suited to be built on a crowdsourced model-- there are long traditions of people gathering to share advice, stories, and troubles and a mutual help ethic of trying to assist one another in the complicated task of making families work.

Tuesday, May 12, 2009

Models of Crowdsourcing Applied to Family Life Education

In his book, Crowdsourcing, Jeff Howe, identifies four general models of crowdsourcing-- crowd wisdom, crowd creation, crowd voting and crowd funding. Each of these models can be applied to the creation of family life education websites.

Crowding sourcing wisdom about family life.

Although we have significant information about human development and families, there are still many applications to particular challenges or particular children that reside in the daily experience of individual parents and family members that could be helpful to others. Most family life educators who work with groups (F2F or online) know the value of inviting the participants to share their strategies and ideas about questions and issues facing one another. In open social networking sites for parents you see a lot of this type of discussion. One parent poses a challenge and others suggest ways to deal with the situation. Sometimes bad advice is offered, but often times there are helpful suggestions. There are many ways to extend and encourage this crowd sourced wisdom from family members.

Crowd creation of family life educational experiences.

Family life education sites can move beyond simply capturing the wisdom of family members and involve families in designing and developing the educational experiences. Family members could be included in serving as a moderator of open forums of parent discussions. Participants might monitor topics of interest to particular families (for example, parents of children with autism) to identify hot, relevant, or new issues. Participants could be invited to write, record or video content that to illustrate a particular point. (Note: There are many developmental issues in the lives of children that can most easily be illustrated by video better than words. Family life education would be powerfully advanced by having easy access to short video clips of these developmental milestones. Asking parents to provide video examples of developmental milestones would dramatically increase the our ability to help parents understand human development and enrich the text descriptions of these topics.)

Crowd voting in family life education.

The simple version of "crowd voting" is to ask readers of family life education websites to rate articles, videos, etc. on usefulness or other qualities. Participants can also be asked to write reviews or reactions to topics. (this may be more crowd creation than voting.) Clearly, if a family life educational website were successful in gathering crowd created material, there would be many opportunities to include the participants in rating and commenting on the various creations.

Crowd funding of family life education.

At present the most common model of funding on the Internet has been an advertiser model. There are still relatively few examples in which people contribute to the funding of content delivery. One model that might work for family life education is the model used by ESPN and tried by several newspapers in which much of the content is available for free, but their is some "in-depth content" created by the most popular commentators that is only available by subscription. This might work in family life education settings. Another version of this would be to offer more individualized experiences for participants that would provide more in-depth support or help through a paid subscription process.


There are numerous opportunities to engage parents and other family members in "crowd-based" strategies for interacting, developing and advancing family life education. This is an important area of further exploration.

Sunday, May 10, 2009

CrowdSourcing Higher Education: The Sturgeon's Law Impasse

One of the major reasons that educators are resistant to opening up education so that many people can participate in the creation of learning experiences is the worry that most of the material created will be badly done.

It is not that educators see their own work as perfect, but they remain skeptical that "committed amateur educators" would produce high quality learning experiences. This concern, sometimes called Sturgeon's Law or Revelation, holds that 90% of the material created by individuals is very low quality and at best 10% of the material will be useful and of high quality. Based on this idea, most educators are not willing to risk trying to find the 10% of valuable material in the face of the 90% of worthless material. They are also concerned that the worthless material will damage the the reputation of their good material.

In designing in open education system that allows for many people to develop content and learning experiences, this issue will have to be faced and a system designed to deal with this problem.

Is there a model that would change educators' views on this issue? Are their tools that would provide an easy way to sift through the material to find the 10%?

Friday, May 08, 2009

Early Learning Educational Platform-- Required Features

Here is a working list of the features and tools that would be available in a robust early learning educational platform.

  • Need to create an educational and technical framework within which a variety of short and extended educational experiences can be incorporated.
  • Needs to have a sustainable development and maintenance system in which people can contribute for short periods and be replaced without major disruption to the enterprise.
  • Needs to be scalable so that the work can grow and yet still be managed.
  • Needs to take advantage of existing material and accommodate new material.
  • Needs to be able to handle text, audio and video formats…and any new formats.
  • Needs to be able to use a variety of levels of manpower in effective ways—interested amateurs, county level extension staff, non-extension professionals, state-level extension staff, university faculty without extension appointments, undergraduate and graduate students.
  • Includes convenient ways for people to contribute individually without much much efforts (e.g., Wikipedia model—in which when you encounter any page, you can register and contribute).
  • Includes a range of short (e.g., text FAQs, 1 minute video or audio clips, etc.) and extended educational experiences
  • Should include educational experiences that serve a range of types of learners from one-time, specific questions to in-depth experiences that would result in college credit. The platform would include all the levels in between.
  • Needs to have ways for contributors to get credit.
  • Needs to be designed in ways that foster credibility with the audiences. This might be different at different levels of the educational experiences.
  • For each type of “contribution” there needs to be easy tools to use to contribute. For example, there are just a few steps to follow to upload a video onto YouTube. Likewise, on Wikipedia, the text editor is right there to use.
  • There needs to be a variety of instructional tools—a text FAQ maker, a quiz maker, a tool to build an educational path that links a series of FAQs into a longer educational experience (from text FAQ to a factsheet that combines several FAQs to a series of factsheets that might be the equivalent of a “book chapter,” a series of chapters that might compose a book or course text.
  • Other instructional tools would be a quiz maker that might use the FAQs, a data collection tool such as survey maker for collecting information that may serve as a variety of feedback, educational and scientific purposes, data presentation tools or ways to easily display charts and graphs, presentation tools or ways to incorporate audio & slides, or text and slides, or video, and probably more.
  • Another instructional tool might be a “story-telling” tool that fosters the development of richer examples of understanding human development and family life concepts.
  • This platform should include a variety of opportunities for social networking and community-building.
  • What else?

Tuesday, May 05, 2009

Crowdsourcing Parent Education

Most professionals who are engaged in teaching parent education and other family-related issues are generally reluctant to embrace the crowdsourcing idea of working with parents and family members.

First, although few people would claim to be experts on physics, chemistry, engineering, software development, etc., almost everyone thinks they are are an expert on raising children and managing families. Second, although many people would agree that there physics, chemistry, etc. are based on science, fewer would agree that behavioral or social science information is much better than commonsense.

So within this context, should behavioral and social scientists engage in developing crowdsourced educational activities in which ordinary people have a chance to create and exchange knowledge and insight about parenting and family matters? This leads to a whole series of questions and puzzles about how do you manage the potential of inaccurate information. What does this do to the credibility of our scientific knowledge about parenting and human development when "untrained" people are allowed to provide insights and advice about parenting, etc?

What risks do you run of being the source of damaging or very inappropriate ideas?

Despite the risks, I am on the side that professionals should invite the "crowd" into the creation of education for parents and families. Although there is potential for misinformation, I think there is much more promise of wise, thoughtful information. With the use of "moderators" and other techniques, misinformation can be a way of correcting parents ideas about various issues rather than treat this information as a problem.

At present we generally don't know the extent to which misinformation is common in forums, chatrooms and social network sites for parents. I know of no efforts to examine these sites to see the degree to which accurate or inaccurate information is being exchanged.

Before we abandon the idea of crowdsourcing parent education, let's see what is really happening.

Sunday, May 03, 2009

Crowd Sourcing Higher Education

Despite all the efforts to begin fashioning open higher education models (see MIT, OpenCourseWare, etc.), there is less discussion of "crowdsourcing" higher education. The basic idea of crowdsourcing is inviting your customers into the business and sharing their ideas and activities within the enterprise.

In higher education I think we are more reluctant to give up our positions as "experts" to our students. Even the various "guide on the side" ideas about teaching in higher education never allow the faculty member to be in any other place than the nominal head of the class.

Yet it seems to make sense to begin including students into the instructional process. There is a considerable body of educational scholarship that suggests that "peer tutoring" can be a powerful force for learning, both for the peer being tutored as well as the tutor. Any what about an even large pool of students, former students and other interested "amateurs" who are interested in both learning and thinking carefully about the content of higher education. Isn't this a great untapped resource? Rather than let them devote all their energy to creating thoughtful Wikipedia entries, shouldn't we invite into the process of developing instructional materials?

Saturday, May 02, 2009

eLearning Infrastructure: A Model

(Note: This continues the development of my ideas about an early learning educational platform.)

Building a learning community not only requires us to think about how we structure the interaction, we also have to think about how we structure the learning. In these figures I conceptualize a model for how learning might be structured in elearning settings.

The general idea in this model is that learning is structured from quick solutions or answers to more complex learning experiences that engage people in problem-solving and deeper explorations of ideas.

Next, I have tried to think about the continuum of learning experiences that would fit into this framework. In this second figure I have tried to identify the range of instructional methods that would fit into this learning infrastructure.

The most elemental forms of learning would seem to be questions and answers. Online this is often referred to as "frequently asked questions." (See previous discussion of "FAQS.") This format has been used a a variety of learning situations and seems like a good place to begin building a learning structure. FAQs do not have to always be text. There can be audio and video "answers" to questions just as easily as there can be "text" answers to questions. The next step in this structure would be increasing the length or depth of answers to questions in a sequence. A simple version of this idea would be a short article composed of a string of FAQs that would describe a more complex topic. Next, I use the term "microlearning" activities to refer to simple interactions with learners. Here I am thinking about quizzes, surveys, true-false tests, and so forth.

The next stage begins to put learners in each other's company so that they begin to learn together. The previous material is generally designed as individual learning. I have chosen the term "peer discussions" for this next phase. At this moment in time the discussions would probably take place in social networking sites in which participants would be invited to talk to each other (in a semi-guided fashion) on topics relevant to the educational activity or subject of study.

The final level in this model is a stage in which "feedback" is added to the learning process. Although it is possible for "peer discussions" about a topic to result in feedback to students, this stage explicitly adds the idea that learners are given tasks, activities and assignments in which they will get feedback about their ideas, understanding and mastery of a topic. I have chosen not to refer to this level as "graded and tested" but rather to emphasize "feedback." Although "grades" are often given in situations, the purpose is to provide a "summary" of the feedback.