some conception of "met growth goal previous term" in growth_df?

[almartin82]

am thinking about what kinds of cuts you'd want 'out of the box' with nyt_subgroups. presumably one at-risk group you'd want to surface is kids who didn't grow last term (or perhaps whose sgp was below 33?). trying to think about how to do this.

you could do some dplyr::lag() work, but with the full saturation of growth windows, those dfs aren't exactly in time-series order. I think what we would need to do is to define a mapping of term -> 'last term' and do a join to get growth status 'last term'. eg fall->spring matches prior year fall->spring; but winter -> spring might match fall -> winter.

@chrishaid let's talk about this.

[almartin82]

related to #163