I am conducting a longitudinal study on how athletes’ perceptions of their training environments evolve over a three-year period. The study involves repeated data collection from youth national team athletes at different time points, with potential variations in participant groups at each phase. Structure of the group changes (Not same athletes every time).

My primary goal is to identify:

Given these objectives, what statistical methods would be most appropriate for analyzing such data? Perhaps machine learning models? I am particularly interested in approaches that can handle:

• Longitudinal data with some participant turnover
• Group comparisons (initially selected vs. final stage athletes)
• Identifying key environmental characteristics influencing selection outcomes

I would appreciate any insights, relevant references, or examples of similar studies.