The study has two groups - experimental and control.
Their attitude-behaviour gap is measured at T1, then they read text that is supposed to either reduces the attitude-behaviour gap (experimental condition) or not affect attitude-behaviour gap (control condition). Then their attitude-behaviour gap is measured again at T2.
What statistical test would be used for two independent groups being measured at two times, once before and once after an intervention?
Thank you
The simplest approach would be to create differences scores between T1 and T2, and then do t-Test to see if the amount of change was greater in one group than the other.
If your groups were created via random allocation, you could use ANCOVA. See this BMJ Stats Note.
Article Statistics Notes: Analysing controlled trials with baseline ...
As Bruce Weaver says, the choice is dependent on how people were allocated into groups, but also what inference you are trying to make. There is a lot written on this problem (google Lord's paradox). A good paper to start with is Wainer (1991, Psych Bull), but lots of other papers too (e.g., www.degruyter.com/document/doi/10.1515/edu-2020-0107).
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts