I would put score Now as the response and score Past as a predictor variable so you are modelling change and the include other variables in teh regression model:

see my response to this

https://www.researchgate.net/post/Is_there_a_difference_between_ANOVA_analysis_of_change_scores_and_repeated-measures_ANOVA

I would put score Now as the response and score Past as a predictor variable so you are modelling change and the include other variables in teh regression model:

see my response to this

https://www.researchgate.net/post/Is_there_a_difference_between_ANOVA_analysis_of_change_scores_and_repeated-measures_ANOVA

Thanks Prof Kelvyn Jones for your suggestion, we always learn from your discussions and responses.

There is a large literature on the choice between running the analyses on gain scores or covarying out the initial score. The choice depends on a few things, like whether the predicted values are if there are no group differences and also whether the groups are influenced by the baseline score. Worth googling "Lord's Paradox." I'll attach a recent review. The papers by Holland and Rubin, Wainer and Brown, Pearl, and the original descriptions by Lord are all useful.

NumPy is an excellent library of Python to process tabular data (if your data are in tables). Python is an object-oriented high-level language: it means, it is easier to learn comparing to other languages (like C). Python also has lots of libraries for plotting and visualization your graphs. R is also effective language for data processing.