Fundamental count dissonance due to overlapping individuals in multi-collection study

[WValenti]

Studies 76 and 77 cover multiple collections and multiple table_ids all grouped under one table_set_id. Unfortunately individuals overlap, which means that for a variable like ADOPTED, there are 3 variables for table_set_id 120, each with distinct (but hopefully identical) answers for the overlapping ind_ids.

This is going to make counts weird, but there's nothing I can do about it in any reasonable amount of time because everything is built around table_set_id, and even if it weren't, we'd have multiple answers anyhow -- they'd just not screw-up summaries so readily.

This is tabled for WAY LATER.