Skip to main content

Body Estimation

Introduction​

Often, in your education so far, you have been required to get precise, mathematically-correct answers with a strong focus on accuracy. Of course, accuracy matters for engineering, but estimation, approximation, and managing error are much more important than 100% correct answers.

For this exercise, you are asked to use only your body as a measuring stick. No using rulers or any other device that will give you an accurate, pre-measured scientific unit. Any part of your actual body is fine.

Materials​

  • Some things to measure
  • A sketchbook to record your notes in

Activity​

Make some measurements​

Measure four different sizes of things:

  • Something small, as in, smaller than your hand.
  • Something medium, as in, bigger than your hand but smaller than your whole body.
  • Something large, as in, something bigger than your body.
  • Something very large, as in, as big as a room.

For each measurement, estimate using only your body parts, and include an error estimate. Start with a rough estimate for each, then refine if time.

Error​

While you're measuring, estimate the error by setting an upper and lower bound.

For example, if I am measuring my pencil with my index finger, I measure it to be 12 index fingers +/- 3 index fingers. I might have accidentally overlapped or spaced out my measuring spots, so it could be as long as 15 index fingers at an upper bound, or as little as 9 as a lower bound. The amount of error for each time I put my finger down is at most 1/4 (+/- 1/8) of my finger if I'm being careful, so out of 12 times of putting my finger down, I could have potentially added or removed 3 fingers.

Accuracy vs. precision​

Consider the difference in error when you use something large, like your hand, vs. something small, like your pinky, to measure the different sizes of other things. Imagine if you had to measure multiple times: what would give you the best results in terms of accuracy and precision.

  • Accuracy: Accuracy is the distance your measurement is from the true dimension of the object you're measuring, where smaller is more accurate. A highly-accurate measurement system will produce an average measurement that overlaps with the true measurement.

  • Precision: Precision is the magnitude of measurement that you're capable of making, where smaller is more precise. A highly-precise measurement system will produce a tight cluster of measurements.

Approximating​

Repeatedly estimating with smaller and smaller error bounds is the process of approximating. Robots commonly use approximating techniques to manage error, and make the best guess about their environments.

On your own​

Memorize your body's measurements​

Being able to quickly and accurately approximate a measurement using your own body is a designer super power. Find a part of your body that is equivalent to each unit you might use regularly.

For example, I know that I am 6ft tall, my thumb is about 1 inch, my pinky is about 1cm, my index finger to thumb spans 6inches, and my pinky finger to thumb spans 8inches (yes, typical Canadian using both SI and Imperial units). I can use this to estimate the size of pretty much everything that I encounter in normal life.

Beware of vanity sizing. I wear pants with supposedly a 33 in waist, but if I measure around where my pants actually sit on my hips, it's 38 inches. Big error if I were to use my belt to measure something!

Philosophical connection​

Ontology (the study of being) and epistemology (the study of knowledge) are tightly linked.

In terms of ontology, most scientists are realists. Realists contend that there is a real world that we can come to know through objective observation. Subjectivists contend that our own perceptions are the only "measurement" of reality that we get access to, so there is no objective world to contend with. Of course, then there are many positions in-between.

In terms of epistemology, what we think of a classical scientific stace is positivism. Positivists believe that we can deduce reality from axioms and observations, usually with an incorporated post-positivist critique that we are often biased observers, so we can only ever deduce reality from statistical measurements, using better and better approximations depending on the size of our sample and precision of our measuring devices.

How does the body estimation exercise challenge or support these philosophical positions? On the one hand, you're measuring the real world. On the other hand, you are using your body as the measuring stick, so all of your measurements are relative to your own body and perception. Discuss with your group.