Data is not the Truth
Bill Shander
Keynote speaker, workshop leader, LinkedIn Learning Instructor focused on information design, data storytelling & visualization, and creativity
Data is not the Truth.
Data is not objective.
Data is not the end-all be-all solution to all of our problems.
But first and foremost, I'll say it again. Data is not the truth.
Data is the result of flawed humans making all kinds of decisions that result in that data. We decide what to measure, how to measure it, what the thresholds or benchmarks are that we then use to consider whether that data is "high or low" or "good or bad" and what it all means. We also define what we consider "accurate" and "precise" for one set of data, which might be very different from another. Sometimes we look at a rate, other times a raw value. Sometimes we're using statistical methods, other times we just look directly at the numbers themselves.
And all of this, as I mentioned at the beginning of that paragraph, happens in the context of all of our flaws. We make all kinds of mistakes in judgment and action. For instance, we are chock-full of biases that affect how we think about literally everything. So any data, which is created by humans, is inherently NOT the Truth.
For instance, I have a thermometer outside my kitchen window. It is in the shade most of the day, so when I consult it to figure out if I need a jacket for a walk around the neighborhood, I get the "in the shade, protected from the wind (usually) temperature", which is not at all the same as the "when I'm walking down John Matthew Rd and the sun is blasting me and the wind is blocked by the hill real-feel temperature", which is part of what I need. It's also not the "walking down Gina Rd into the stiff wind that always blows up that hill temperature" either.
Recommended by LinkedIn
So it is not the "Truth".
All of this is just a reminder (to self, as much as to you) that data is not perfect. Data is flawed. It may be the best thing we have in some cases, but it is not everything.
This summer I'll be shooting more episodes of my Lesson & Listen series and one of the topics is about Data Equity where I'll be talking a lot more about this (and interviewing an expert in the field, as always)!
==========
Ask me about virtual or in-person data storytelling and visualization training for your team
Learn more about data storytelling and visualization via my other LinkedIn Learning courses
Semi-Retired
1yData is just data. It's accuracy, completeness, reliability, consistency and meaning is open to to debate and examination. A saying, "Figures lie and liars figure" is another to assess data's truthfulness.
Data Scientist / Machine Learning Engineer
1yIt truly works both ways. You are spot on about datasets with missing critical attributes, those which are poorly modelled, or otherwise flawed. However, there are also situations where the data is accurate but people aren't willing to explore the possibility that their processes could be improved or deeply rooted biases could be wrong.
Editor, proofreader, and sensitivity reader for indie authors at The Assist LLC. Stop throwing Hail Marys and write confidently.
1yI love this. So many get stuck thinking their data paints the entire picture, when in reality the data is just one component of the story. Such an important reminder!
—-
1yBut art is! 🕊 "Art is a lie that makes us realize the truth." - Pablo Picasso