Hi everyone,

I conducted an experiment where 2 parameters (alpha, beta) using 6 different methods were estimated for each participant (number of participants 300). The final values of these parameters are their mean values (e.g. mean_alpha and mean_beta estimated by method 1, mean_alpha and mean_delta estimated by methd 2, ect.). The same survey with the same participants was repeated after one month to see if the results are time-consistent. This time only 250 participants out of the 300 participated. I am not sure about two issues:
  1. the best way to present the data of each participant in time period 1 and time period 2.
I have organized the data like this, but I am not sure if this is a good way, especially for further analysis:
User Method Mean_alpha_period1 Mean_beta_period1 Mean_alpha_period2 Mean_beta_period2
1 1 1.05 0.88 0.77 1.21
1 2 1.60 1.24 0.62 1.03
1 3 0.9 1.10 0.83 0.72
2. to check if there is a difference between the methods in estimating the parameters alpha and beta with different methods, I conducted a paired ttest (ttest mean_alpha1== mean_alpha2) separately for data in period1 and separately in period2.
The idea is to get the same values for the 2 parameters using the 6 methods, so there should be no difference in estimating alpha and beta with respect to the method.

Now I have to check if the parameters estimated by the six methods are different after one month for each participant. I thought to use difference-in-difference, but I am not sure if this would be appropriate. If yes, would the code then be:
didregress (mean_alpha_period1) (method)

Thanks.