Lately, I've come to the realization that I'm not very good at running experiments. I'm good at what I have come to call "spaghetti." Spaghetti is where you throw things at the wall and see what sticks... if you want to see where the term comes from, give a 2-year-old spaghetti for dinner, don't watch him too closely and see what happens when he throws the spaghetti against the wall. Don't get me wrong, sometimes spaghetti can be useful. However, it's not an experiment.
Let me give you an example. I've been treating this blog like spaghetti. I've been trying some neat things, but not coming up with disciplined experiments. I fiddle around with an idea or just try some stuff, but there's no hypothesis. This makes gathering feedback or developing measures tough because I'm not clear about what I'm testing. Fortunately, I've got people around me pushing me on this, which has helped me realize my spaghetti mistake.
With all that said (er, typed), I want to run experiments and stop throwing spaghetti. I need your help with this. When is something an experiment? Moreover, how do you turn spaghetti into a useful experiment? Specifically, how can I turn my "blog spaghetti" into a useful experiment? Help me out in the comments.
One experiment: which "call to action" in a blog post yields the most responses?
Posted by: Peter | 01 November 2010 at 04:58 PM
Keep these things in mind as they are crucial to the scientific method:
Observable
Measurable
Repeatable
When you plan an experiment, don't expect to get more out of it than just a "hmm... that's interesting" unless you create something that is measurable and repeatable.
Depending on your intended result, observable should be pretty simple for a blog. (not to be confused with easy)
Posted by: David McGinnis | 02 November 2010 at 09:51 AM
David- I like the three attributes. It seems like I could write those out how my experiment fits into those to see if my experiment makes sense before trying to put it in place.
Peter- Thanks for the thought about potential measures. Would just counting comments be enough evidence?
Posted by: Ann Zerkle | 02 November 2010 at 01:28 PM