According to Kruschke (http://www.users.csbsju.edu/~mgass/robert.pdf), if I have two different groups and collect their response times to a certain task, to determine if the two groups are significantly different I would take the difference in posteriors and examine the 95% highest density interval, and then see if it overlaps with zero (or with the ROPE). But if I had a single group contribute response times for each of two conditions, it seems that you do the same thing -- subtracting the posteriors and examining the HDI. But I thought that paired samples would afford greater statistical power, just like in frequentist statistics. Am I missing something?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.