Posted by: bbannan | February 25, 2009

Representations of Data – Cassandra

In his book, Visual Explanations, Edward Tufte offers a classic example of how various representations of the same data can lead to extremely different conclusions. (Read his example “The Cholera Epidemic in London, 1854” [ Decisions.pdf] pgs 2-7.)

After reading Tufte’s analysis, consider the various ways this same information may have been misconstrued. What was different about Snow’s approach? How might other representations of this data have altered his conclusions? What action(s) might be taken to avoid misleading interpretations of raw data?

In the comments below, offer any example from your life where representation altered your interpretation of data. How did this representation change how this data was perceived, and what impact did it have (on your or the project)?



  1. Tufte’s analysis followed a degree of disciplined thinking. The handle was not removed before he had carefully thought through correlations of data, then tested to determine if there might be cause and effect linkages.

    His initial research question caused him to ask what other factors correlated with the cholera outbreak. A visual depiction of deaths on a map enabled him to see the pump within the disease outbreak area.

    That was valuable, but perhaps more valuable were his findings from the brewery where the water was not consumed and provided negative indicators for cholera. In effect this enabled him to test by removing one variable from the other and seeing a significantly different outcome.

    The brewery observations might have been interpreted differently. The deduction could have been drawn that drinking beer inoculated against the disease.

    In effect, by removing the pump handle he was attempting to replicate the brewery findings.

    We all make observations, collect data, perform some level of analysis and then reach conclusions. At times we speed up the process by applying mental models, checking to see if the information fits the model. An example of this is my experiences driving in snow and on ice. As a young man I learned to drive with non-power assisted brakes. When driving on slick surfaces sometimes the car would break traction and the front end begin to slide right or left with the reverse effect to the rear. The automatic response was to turn in the opposite direction (if I recall correctly) of the front end of the car.

    With the advent of anti-lock brakes this is no longer necessary. But getting used to anti-lock brakes in these type situations took time and a conscious effort to not apply my manual brake mental model.

    The important consideration to draw from application of mental models is that they often outlive their usefulness. They do speed up our decision-making process. But by applying mental models we tend at the same time to suspend our reasoning process.

  2. Cassandra, and Hank
    I read the material and thought about it for some time. There was a good reason for my wait. After reading Hank’s response, I could not think of a way to say the same, because I agree with his statement, without simply copying what he had said.

    We are, in fact, the sum of our thoughts. As those thoughts are found, through observation or experiment, to be less helpful than in the past, we make amendments to them, testing new plausible outcomes and attempting to determine if these are a better fit to produce the desired results.

    It is important to realize that nothing is static, whether it is the process or our interpretation of it. Knowledge in itself is not enough; there must be constant testing of that knowledge against changing circumstances if we are to consider our perceptions viable.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: