In a previous post, I discussed information as a pyramid of human communication, that starts with data and ends with wisdom. So, how do we turn data into information? Information into knowledge? The work that MDM students do plays a key role in both of these transitions. In this post I will explore the means by which we turn data into information and information into knowledge.
How Do We Create Information?
My thinking in this has been strongly influenced by Thomas Davenport and Larry Prusak, who wrote Working Knowlege, (HBS Press, 1998). Their book is still the best thing I have read on how people and organizations can turn data into information and information into knowledge.
Davenport and Prusak first point out that data is all around us and that it is essentially "what is." While that may be true, it is helpful to at least put a bit of a ring around a data set and suggest that we have data about something. At the time they wrote, the term "big data" didn’t yet exist but it is good to remember that whatever data set we have, it is typically at least somewhat contained. Perhaps it is how people are using a mobile app, and we have this data in the form of clicks on things. Or it could be a list of names and numbers. It could be data from vehicle sensors, as we will see below.
From Data to Information
For this data to become useful (their definition of information is data that is organized somehow) we have to do something to it. It needs to be transformed. Davenport and Prusak suggest that there are "5 Cs" to how we might do that. These are:
- How is the data contextualized? Do we know why the data was gathered?
- How was the data been categorized? Do we know the units of analysis, the key components of the data?
- How was the data calculated? Have there been some mathematical or statistical analysis, such as changes over time, averages, etc?
- What corrections have been applied to the data? Do we know how and whether or not errors have been removed?
- And finally, has the data been condensed? Are there summaries, tables, graphics?
If you take something familiar, like a phone book (well, it used to be familiar!), then you can see how data in the form of names and numbers is turned into useful information by context ("It is a phone book for Vancouver, created in 2012."), by categorization (Residential and Business listings separate and sorted alphabetically, perhaps sub-sections for Yellow Pages groupings, headings), and correction (ensuring by hand and perhaps by machine testing that the numbers are all working and made up of 10 digits). Calculation and condensation are not typically major aspects of a phone book, but they might be major elements of a book summarizing a baseball season or productivity in a manufacturing plant.
Creating information out of data is a worthy pursuit, and adds considerable value to what would otherwise be something pretty difficult to use or irrelevant (imagine a phone book without alphabetization). But that typically does not lead to knowledge. It doesn’t (yet) help you with a decision. Who should you call first about a blocked toilet, for example?
From Information to Knowledge
Turning information into action is the next step up the pyramid and what defines knowledge. Some call knowledge "actionable information." Another transformation is called for: Davenport and Prusak helpfully provide another list (and another set of "Cs"):
- The information is compared. How does this situation compare to other situations we have been in?
- The consequences are identified. What implications does the information have for decisions and actions?
- Connections have been made. How does this bit of knowledge relate to others?
- A conversation is initiated. What do other people think about this information?
If I look back over the past few years for examples of MDM student projects that are turning data into information, and even information into knowledge, I can find quite a few:
- Turning sensor data about air quality into advice for runners, cyclists, asthmatics: AwAir
- Creating game interfaces from new motion sensing input devices: Pix
- Designing new ways of interacting with a sustainability expert system: Go/IDEASS
The trick is recognizing how we are doing this, why we are doing it, and then appreciating the importance of that task. Especially the responsibility that exists when you turn data into useful information and information into actionable knowledge.
Watch Student Team Go Explain How They Approached The IDEASS Sustainability Planning Project
Davenport, Thomas H., and Lawrence Prusak. 1998. Working Knowledge: How Organizations Manage What They Know. Cambridge, MA: Harvard Business School Press.