June 9, 2014 (old blog)
Let’s put all our ducks in a row… what statistical methods will we use to analyze the quantitative RIPPLE data? At the moment, this question seems a bit daunting to me; thankfully I am headed to John Hopkins University next week to further develop my statistical reasoning skills.
If you remember from a few weeks back, I blogged about our data collection time points: (1) baseline (when parents complete the RIPPLE intervention in the waiting room) and (2) 1-month follow-up (when parents will be asked to complete the same theory-driven measure that was administered at baseline, which queried intentions, attitudes, motivation, etc.). Now, assuming that some parents will complete the follow-up survey at 1-month (stay tuned for a post on this when I get back!), we would use a specific statistical test (the paired t-test) to compare a parent’s responses at both time points. As well, we can use a similar test (the independent t-test) to compare all parents’ responses at one time-point. Together, these tests will allow us to evaluate both within- and between-subject differences, respectively.
Stay tuned for a new post upon my return in a few weeks… hopefully my statistical skill set will become more sophisticated before then!