## help to decide statistic approach

Hi, I am a newbie in statistics and many thanks in advance.

My experiment involves three subjects and two types of treatments (A, B). Because the treatments only control a symptom (S) in a short period of time, we can repeatedly apply treatments to one subject multiple times and do not worry about carryover effects at all.

The experiment was conducted in this way: For each subjects, we repeated treatment A eight times and treatment B eight times (sequence randomized). We measure the symtom (S) before each treatment and after each treatment. The question that we want to answer: 1) does treatment A or B affect S? 2) Is improvement on S the same for treatment A and treatment B?

We were thinking different options:
1) two way anova,
2) repeated measurement anova,
3) paired t-test for each subject first and paired t-test for average improvement of S for all three subjects.

Would you mind recommending any of them or anything beyond?

Thank you so much.

Best.

Ming