Choose Your Own Data-Analytic Adventure

bob.8
So what should I write in the IRB proposal'

First, since you randomized the intervention and control groups, they are equal (in expectation). Therefore, if you observe a statistically significant difference in anxiety, you can reasonably infer that the difference was caused by your child life intervention.

Conduct a regression analysis. Regress your continuous outcome (MYPAS) on your dichotomous predictor (Intervention). In other words, fit the following model:


model


Estimated ß1 will be the average intervention/control anxiety difference in the sample, which is an unbiased estimate of the average intervention effect in the population.


Conduct a t-test to determine if estimated ß1 is statistically significantly different from zero. In other words, test the null hypothesis that there is no average difference in the population between anxiety in the intervention group and anxiety in the control group. If you reject the null hypothesis, conclude that there is an average difference in the population (and that your intervention caused that difference).

Use 95% confidence intervals to gauge the magnitude of the average intervention effect in the population.


You reply:
"That's beyond me. Good luck. Let me know how it goes."
or
"Statistical power' Sample size' We can give it a shot."
or
"Let's reconsider how you will characterize the outcome."