how raw are our data?

[davidwhogg]

Here is the issue I am concerned about:

• Many exoplanet projects report only planets, never non-planets.
• Even those that tell you where they didn't see planets don't give useful limit or measurement information.
• Is our data source "the literaure" or, say, "the pixels" in the data?

I think for some data sets, we will be able to see all the measurements and project models into the measurement space, but for other data sets, we will only be able to go on the claimed detections in the literature. This has to somehow be part of the model.

This is a problem and an opportunity, of course!

[davidwhogg]

ps. I am thinking of this like (a) completeness in PHAT (there is a p(observed | stuff) or in this case p(reported | stuff) ) or like (b) Comet Holmes (there is a p(observer did x | stuff) ).