KM in Education

Data for schools and classrooms

Collecting and analyzing the tremendous amount of data required for No Child Left Behind has prompted school districts and states to heavily invest in data management solutions. It is one of the fastest growing spending areas in education spending, with Eduventures estimating a market value of $145 million in 2003.

Data warehouses have emerged as one of the key data management components in education agencies. By definition, these computer based information systems house "secondhand" data originating from other data sources and applications. Warehouses facilitate better decision-making by enabling more robust data analysis and reporting using data from multiple sources.

With data warehouses, education administrators can analyze relationships among different data sources such as assessment tools and student information databases. School programs could be evaluated by linking participation data with performance data. Schools may also find a link between professional development and performance data. Effective data analysis ultimately will help schools make smarter decisions.

For teachers, data marts, or subsets of data warehouses, offers the next step to getting the data they want for their courses and their students. Some products like IBM's eScholar are now exploring how to help teachers access this information via handhelds. Imagine how easily a language arts teacher could use her handheld to identify which students have below standard assessment scores in English, so she can provide them with extra help.

The real question is how long will it be before we get to that point where all educators have ready access and the ability to obtain such information?

Technical coordination

The latest knowledge efforts in public education have been dictated by government regulations such as the No Child Left Behind Act (NCLB). School processes are changing to meet new NCLB requirements and, in the process, have had to often upgrade their technical capabilities. For example, before NCLB, longitudinal student data was only collected by a handful of states. Now, about half of the states are equipped with systems to collect longitudinal student data, mainly to meet NCLB and state reporting requirements. A 2002 policy brief published by the Education Commission of the States provides a clear explanation of how NCLB would impact state information systems.

By upgrading their processes and relevant systems, states have increased their technical capacity for coordinating different sources of knowledge. Linking attendance data with performance data at the individual level creates a more complete picture of a student, empowering schools with new knowledge. However, with these upgrades, many districts have also found themselves grappling with different types of systems and even more data and content.

In my next posting I will briefly describe a couple of these systems and the challenges they pose.

Student-focused processes

As a former process consultant, I have sometimes been overwhelmed by the number of process improvement opportunities I have observed while visiting a school or speaking with an educator. One opportunity that frequently strikes me is building processes around students.

I like to think of students as the ultimate customer for educators. A customer focus has long been a critical component to successful business processes. Yet despite the focus on student outcomes, educational institutions seem to tailor few of their processes around students.

If public education had focused on students, the process of creating a student record would have long ago incorporated the longitudinal history of a student as s/he developed academically, emotionally and personally across grades, schools, districts, and even states. A holistic student focus would have also gathered knowledge from all facets of a student's life, from home, extracurriculars, and academic enrichment activities, to build a complete picture of a student's situation. Armed with this knowledge, a teacher would be better prepared to design lesson plans to address students' specific strengths and weaknesses, to be sensitive to extraordinary family situations, and to suggest appropriate enrichment activities not tried before.

Student-focused processes also have tremendous potential at the aggregate level. Just as customer surveys provide valuable input, student feedback could help schools improve their inquiry processes toward continuous improvement. Student performance could also serve as input into curriculum design.

In summary, most school processes still acknowledge a gap in fully addressing student needs. Schools are closing that gap by slowly incorporating more knowledge from and about students at both the individual and aggregate level. It is in this transition that schools are discovering technology as an enabler for coordinating knowledge.

Coordinating student knowledge

Amazon's website demonstrates the brilliant marketing that has evolved due to increasingly sophisticated coordination of vast amounts of customer knowledge. Its ability to correctly identify new products or books that a customer might like to purchase exemplifies the private sector's capacity for effectively targeting customers by capturing and integrating customer knowledge into comprehensive customer profiles.

Education is headed in the same direction, toward integrating knowledge about each student, but is still far from using the knowledge to effectively help student achievement. From a technology standpoint, integration of different student information sources has only been enhanced in the past few years by the adoption of the Schools Interoperability Framework (SIF) that enables K-12 software applications to share data. From a process standpoint, however, schools are still far from fully integrating different knowledge sources to create a complete student profile. For example, what one teacher has learned about a particular student over one year is rarely passed on to the student's teacher in the following year.

Pulling together all the knowledge about a student entails not only collecting all the student's performance data in the form of assessment scores, grades, and attendance, but also capturing all the observations and learnings from each adult relationship, whether it is a teacher, guidance counselor, parent, or afterschool tutor. Only then, will school staff, armed with the full context of an individual student's situation, be able to make effective decisions for improving an individual student's achievement.

My next postings will address in more depth how coordination is evolving at both a technological and process level.

Youth Voices

An interesting conversation today with Jonathan London at Youth in Focus raised the idea of youth knowledge becoming an input to more school processes, as well as at the policy level. Youth in Focus helps schools and community-based agencies to enable young people to engage in action research and problem-solving to improve broader communities and/or institutions. The resulting youth-produced knowledge can be used as input to school improvement, curriculum development, or even state policy decisions.

Independent school networks

U.S. independent schools, by virtue of their smaller networks, are in a great position to tap into the resources of similar schools around the country. I spoke today with Ed Hu at Harvard Westlake in Los Angeles, and was struck, for example, by how easily he could contact a school of like size and culture in New England to request advice for starting a capital campaign. Independent schools know each other well, and the staff from such schools quickly develop informal networks across the country from attending conferences and years of working in the sector.

How can we help traditional public schools and their staff reach the same point? Where they can easily draw upon the expertise and experience of schools like them? Some education support organizations facilitate such connections, but otherwise, schools operate largely in their own spheres. Sometimes they do reach out as one teacher from Seattle shared with me. Her school had identified a similar urban school in California with demonstrated success in student achievement, then funded a trip for selective faculty to learn from the school's practices, only to find that the school embraced a different culture and their practices were therefore not appropriate for the Seattle school.

My suggestion is to look more closely at how independent schools form their networks and, if possible, model similar processes for traditional public schools.

Collaborative Potential

Despite the common goal of improving student achievement, education organizations are often split over the best approach to improving public education. This divergence masks the reality of the situation, that regardless of the school and the approach it embraces, each school has enrolled students for the purpose of learning.

If everyone is truly working in the interest of students, once students are enrolled in schools, it would seem that mutual collaboration across different camps might highlight and enable cooperation in addressing similar challenges. An organization such as the Charter Schools Development Center which provides technical support to charter schools may differ in philosophy from the Coalition of Essential Schools, which emphasizes the development and sustainability of equitable, intellectually vibrant, personalized schools. Yet they both highly value the dissemination of knowledge to their schools and probably face similar challenges in keeping information fresh, relevant, and protected from copying. Given the dearth of resources to devote to knowledge dissemination, working together to address such issues may result in stronger processes and tools for both organizations, and ultimately better practices toward the education of each student.

Classroom web logs

Today's New York Times article on blogs raises some interesting points. I challenge why some disappointing blogging results have:

led some teachers who are critical of blogs to question whether the technology has actually done anything to interest students in writing. Critics also worry that the casual nature of writing on the Web may encourage bad habits that are hard to break, like e-mail-style abbreviations, bad grammar and poor spelling.

Technology has never been a solution for learning and should not be relied upon for motivating students to learn. It is rather the content or process of the lesson that should be interesting, and further facilitated by collaborative technologies such as web logs!

Embedding knowledge

Even with the capacity and ability to organize its knowledge, an organization does not ensure its ability to effectively leverage and apply its knowledge.

As I have been helping an Oakland middle school redesign its afterschool program, it has become clear that communication is a key element to be improved. Teachers and afterschool tutors both hold critical knowledge about students, yet this knowledge is not effectively shared between the two groups. By using different tutoring approaches and knowledge capture strategies, tutors can organize knowledge so it is more accessible for teachers. Yet it still remains useless unless teachers take the time to access it and use the knowledge.

We recognize that it is unlikely that busy teachers are going to proactively check for new knowledge, whether with the tutor (tacit) or from documentation (explicit). It would become one more task in their already bulging days. However, teachers do always check their mail or in-boxes, and it is a relatively simple matter to embed critical afterschool tutor knowledge in that daily check-in. Accessing this new knowledge is now a natural extension of a teachers' habits, and the teacher is much more likely to act upon the knowledge s/he has received.

Organizing knowledge - Part III

Taxonomies and meta data are simply tools for organizing both tacit and explicit knowledge to make them more accessible.

A Rolodex represents a low-tech version of accessing tacit knowledge. An online example would be the Faculty Index at Stanford's School of Education. Research Interest has been used as one type of meta data, and all the different Research Interests make up the basic taxonomy. Because the Stanford faculty agreed on a common term for each Research Interest, anyone who wants to locate all the Stanford faculty associated with a specific interest can do so very easily given this knowledge organization method.

Examples of explicit knowledge organization abound. However, organizational approaches still remain rather simple. The GEM website uses basic metadata such as subject, type, level, and keywords. Clicking on Browse in the navigation box allows users to view the simple taxonomy associated with the meta data.

By establishing base metadata and taxonomies, organizations reap the benefits of managing knowledge. Internal employees can both share and locate relevant knowledge easily. Sharing across organizations, however, becomes more challenging as the knowledge organization tools are often defined differently from one to another. In the education sector, it is the definition of standards by organizations such as GEM, that will enable greater sharing of knowledge among educators, schools and districts.