dance stats results

you made it!

thanks for being part of the experiment.

Each of the pieces you watched was created using statistical experimental design, a way of strategically setting up experiments to measure as many different things as possible, in as little time and with as few resources as possible. It's great for quality control in manufacturing. I think it's safe to say this is the first (only?) time it's been applied to the art of dance.

We set up the experiment to measure five things. Each of those five things was in each piece at what we're calling a "high level" or a "low level" I like to think of it as a series of five light switches. Each of the 32 dance pieces we made has a unique combination of switches on and off. By giving us your opinion, we can see which of these things you responded to as an audience member -- which switches have the greatest effect on how much you like a piece of dance.

So, what did we test for?

1 CLASSICAL MOVEMENT

The movement of each piece had mostly classical dance shapes and movements (borrowed from ballet, classical modern, and jazz) or focused on more unusual dance shapes with gestural or pedestrian movements.

2 UNISON MOVEMENT

The dancers in each piece mostly moved together in unison or moved individually.

3 RHYTHMIC CLARITY

The dancers moved with precise rhythm and musicality, or felt their way through the movement, choosing their timing over the accompaniment.

4 RELATIONSHIP

The dancers interacted with each other through partnering or working off one another, or moved as independent artists, not depending on the other performers on stage with them.

5 TRADITIONAL ACCOMPANIMENT

This one is about the music. We chose accompaniment for each piece that either matched how the style of dance is traditionally performed, or feels like a less obvious musical choice. If you're finding it hard to wrap you head around this one, think of the macarena. Now think of someone doing the macarena dance to, say.... Mozart. We're calling the Mozart macarena a "non-traditional accompaniment."

SECOND-ORDER INTERACTIONS

This experiment also allowed us to test for second-order interactions, or how two factors together shaped your opinion of a piece. For example, if you responded to an interaction between the first two factors, classical movement and unison movement, you might prefer to see classical movements danced in unison, but quirky gestural movements danced in non-unison -- even though you might not have a preference for either type of movement.

so let me tell you about yourself.

Or, you know, at least your preferences in contemporary dance...today.

The choreographic factors and second-order interactions you responded to most strongly are:

(ps. if this chart is blank, you didn't show any preference strong enough to be statistically significant. You unicorn.)

see an example

Here's an example of a piece that contains elements you responded to. Whether or not you like this particular piece, you may be able to recognize your factors here...

how you stack up

Adding your responses to all responses so far -- both in this online experiment and in our live experiment over four shows in spring 2016:

The choreographic factor audiences respond to most strongly are:

see an example

Here's an example of a piece that contains elements the general audience has responded to. Whether or not you like this particular piece, you may be able to recognize the highlighted factors here...

the math, the research, and the details

Read the full analysis, learn more about statistical experimental design, about the project and what's next.