5.9 Extension activity: Estimating effect size

When observed results are not compatible with the null hypothesis (suggesting that there is an effect), applied researchers often want to know how much more effective the experimental condition is than the control. Statisticians refer to this quantity as an effect size. It is so called because it measures the effectiveness of the experimental condition.

For example, remember in the PEP activities from Week 8, we analyzed a study that investigated the effect of swimming with dolphins on clinical depression.

In this activity, we’ll explore the question,

How much larger is the improvement rate for patients suffering from clinical depression who swim with dolphins than those that don’t?

Notice that we’re not asking about whether there is an effect, we’re asking about how big the effect is. This is a statistical estimation question.

Note, this is an optional extension topic

To access the activty
Log in to your class in Desmos. Find the activty called,
5.9: Extension: Estimating effect size