That's where I think specifically the Active exercises can help. We gave students in a CS2 optional typing practice throughout the semester [1]. These students performed better on exams, made higher learning gains and uploaded less buggy code to GitHub for projects. The results were also significant for students that performed below the median on the first midterm and only started doing the exercises AFTER the midterm. Granted, the students were still self-selecting to improve their status, but it does help establish methods for helping even students in higher level courses.

[1] https://dl.acm.org/doi/pdf/10.1145/3373165.3373177