A couple of weeks ago, talking with Steve Borgatti at the CASOS Summer Institute at Carnegie Mellon, I raised some concerns about how well we are really measuring the phenomena that we describe with our imperfect “tools of measurement”. Steve suggested that my comment revealed a “postmodern” perspective and I will admit to that tendency.
Yesterday, however, while I was reading Daniel Gilbert’s insightful “Stumbling On Happiness” I stumbled upon a great response to the question of subjectivity in the measurement of human experiences. Gilbert addresses this concern by presenting three basic premises:
“The first premise is something that any carpenter could tell you: Imperfect tools are a real pain, but they sure beat pounding nails with your teeth.”
“The second premise is that of all the flawed measures of subjective experience we can take, the honest, real-time report of the attentive individual is the least flawed”
“The third premise is that imperfections in measurement are always a problem, but they are a devastating problem only when we don’t recognize them”
Thanks, I needed that.
One of the advantages of having a diverse group of friends is the opportunity to learn about things you might not normally be exposed to. When I was explaining to my friend, Flann Lippincott, who is a graphic designer, about visual representations of social networks, she said “you must know Edward Tufte”. When I admitted that I didn’t she provided me with a copy of “The Visual Display of Quantitative Information”.
Given our increased ability to create graphical portrayals, Tufte’s “principles of graphical excellence are more relevant than ever. According to Tufte “excellence in statistical graphs consists of complex ideas communicated with clarity, precision, and efficiency.
Tufte by the Numbers
- Show the data
- Induce the viewer to think about the substance rather than about methodology, graphic design, the technology of graphic production, or something else.
- Avoid distorting what the data have to say.
- Present many numbers in a small space.
- Make large data sets coherent.
- Encourage the eye to comparedifferent pieces of data.
- Reveal the data at several levels of detail, from the broad overview to the fine structure.
- Serve a reasonably clear purpose; descriptive, exploration, tabulation, or decoration.
- Be closely integrated with the statistical and verbal descriptions of the data set.
- Graphical excellence is that which gives to the viewer the greatest number of ideas in the shortest time with the least ink and in the smallest space.