Constructing an Experiment

In order to reject the null hypothesis, you collect data from all of your conditions, and then do statistical tests that compare your sets of data.

Do wearable computers help people keep appointments?
How well do people keep appointments without wearables?
They keep appointments about as well either way.
Have some people set up appointments. Count how many keep them.
Do subliminal cues improve memory performance?
How well do people do on the memory task without subliminal cues?
The subliminal cues have no effect on the memory task.
Show some pictures with a subliminal cue, and some without, and count how many are correctly recognized.

Experimental Design for Wearable Computing -- Designing the Experiment

Designing the Experiment

Different questions provoke different experimental designs.

Have some people set up appointments. Count how many keep them.
Are subjects using their own preferred appointment-keeping mechanism, or does the experimenter tell them how to set up the appointment?
Show some pictures with a subliminal cue, and some without, and count how many are correctly recognized.
Which pictures are shown in which condition?