When Collaborating Online, How Much Data Expertise Is Enough?
By Dennis D. McDonald, Ph.D.
How data are shared has an enormous impact on how useful data and analytics will be to any organization. It’s gratifying to see not only how advanced and accessible data analysis and visualization tools have become but also how receptive people are to using these tools.
This is never been truer than today when we need to work remotely. This has forced many organizations--some for the first time--to seriously consider how people can work together when face-to-face meetings are impossible.
In such an environment any impediment to communication can loom large and disrupt the chain of events that must occur in getting data from its source to the eventual user. The more “hoops” the user must jump through, the more likely it becomes that data access and usage problems will arise.
Typical challenges to making data useful that can be exacerbated by having to work remotely include:
CONTEXT. Failure to understand the circumstances surrounding the source of the data.
FRAGMENTATION. Incomplete or fragmented transmission and delivery of the data.
UNDERSTANDING. Lacking a basic understanding of how to analyze and interpret data.
Context
The first of these challenges is the challenge of context. That is, how much do people need to understand the circumstances surrounding the events being described by the data? For example, If the data are part of an ongoing series of reports that track an event or condition over time, are communicated statistics enough for people to understand whether or not an upward or downward trend is occurring? Do they possess the necessary contextual information they need to interpret the significance of the data being communicated?
Compare disease tracking metrics with, say, weather prediction. We all know based on experience that when we get up in the morning we want to know, yes or no, is it going to rain today? But we also know based on experience there is no certainty around weather forecasts. WE learn to live with that uncertainty.
Imagine in comparison how difficult it can be to communicate, in real time, complex data series such as changing COVID-19 infection rates where many people are listening in to a teleconference or online video chat session.
If everyone tries to talk at once, confusion reigns. If only the quick or nimble are able to ask questions, what happens to those reluctant to “raise a hand” electronically? Finally, when many people are online and complex data are being shared how likely is it that a subset of participants will fail--through no fault of their own—to understand the context of the data being communicated collaboratively?
This challenge of context when communicating complex data collaboratively or in an online group setting assumes you’re trying to do more than just broadcast data but are also intent on discussing it collaboratively where, ideally, the thoughts and questions of many people people can be surfaced, communicated, and addressed. Doing this with complex topics is a challenge.
Fragmentation
This has implications for how to plan and manage online collaboration assuming you want to avoid the second challenge: incomplete or fragmented data transmission based on the user not receiving or hearing the whole story.
Anyone tasked with presenting a research paper at a professional conference will understand the challenge of deciding, in a limited time, what to present I want to leave out--and still have time left over for questions.
Add to this the added challenge of presenting a complex research project online to people conferencing in from multiple locations via audio and or video who likely represent a wide range of understanding of the “backstory“ you know is important. In such a situation you need to consider what you need to do to avoid an “…incomplete or fragmented transmission of the data.“ That means controlling your message through planning and by having a good understanding of where your “collaborators“ are coming from. You need to take seriously your responsibility for planning and managing online collaboration (assuming you want to do more than just lecture to a passive audience). Some implications of this:
Control your “data message” as much as possible. Focus on what’s important and what people need to know to make sense of the data.
Use the real time feedback possible with online collaboration technology to make sure your audience “gets it.”
Follow up post-session with online participants, i.e., keep the collaboration going.
Understanding
This brings us to the third challenge: how much do your audience of collaborators need to understand about data and analytics in order for a data-based message to be understood and acted upon?
This question increases in importance exponentially as the size and diversity of the collaborating audience increases. Do you respond to the temptation to appeal to a “lowest common denominator” approach to date in a presentation? Many folks don’t have an even cursory understanding of basic statistics. The wider your audience of participants, the more you need to take that into account in your planning appropriate follow up.
The days when you could publish data files intended for public consumption online and then walk away are long gone. Even simple graphics can be the source of much confusion when used to present data to the general public. For example, how many non-statisticians really understand the simple concept of a “moving average” when used to help illustrate a trend?
Conclusions
Whether the context is education, management, public policy, or research, collaboration regarding data access and usability needs to be planned and managed carefully especially when it has to be done remotely.
Still, I’m not so naïve to as suggest that planning per se is a necessary and sufficient condition for success. You also need to assume (and hope) that chance and serendipity will occur in connection with the data your present.
This is why I believe that knowledge sharing and collaboration are so important when it comes to managing and governing data.
You want to maximize the likelihood that new or unexpected and potentially beneficial ideas arise around the data you share. The more that people understand the data you present and the context of its collection and distribution, the more likely that innovation and creativity will be sparked.
Copyright © 2020 by Dennis D. McDonald
