Implement Josh Yardley's statistics

[matthew-white]

https://ipastorage.box.com/s/0eagp9vr0ttkwxpf8w5t

The Research Rap is here:

https://ipastorage.box.com/s/ze8sri9iano5gm7nan4u

[mbomby]

Just want to re-iterate the importance of this. Josh's important statistics are the Simple Response Variance and Reliability Ratio, which can be estimated from backcheck/ reinterview data under certain assumptions. Integrating into bcstats would definitely facilitate measurement studies within IPA based on these (and possibly other) statistics.

Note that a more extensive review of the theory, ideally in consultation with a statistician, should be done before implementing this.

[boyercb]

I definitely like the idea of making bcstats more analytically rigorous. However, I wonder how our method of directed back check questioning would affect this if at all? Maybe we can do some research and come up with a systematic analysis plan for back checks after GST. With that in hand, I am confident that I could update bcstats in the second quarter of the year.

Chris