In working on markedness, I came to realize that many (perhaps most) of the descriptive work that I relied on are inadequate as sources of evidence for theories of the phonological module.
Descriptions are rarely written with the goals of cognitive science in mind; they often aim to teach, spread religious ideas, or are in an incompatible descriptivist tradition that does not provide appropriate data or evidence for cognitive scientists.
My concerns have broadened to worrying about how to identify and evaluate appropriate evidence.
I recommend starting here:
de Lacy, Paul (forthcoming 2027). Evidence and methods in phonology. In Adam Jardine & Paul de Lacy (eds.)The Cambridge Handbook of Phonology. 2nd Edition. Cambridge University Press.
And here’s another slightly longer article:
de Lacy, Paul (2009). Phonological evidence. In Steve Parker (ed.). Phonological argumentation: Essays on evidence and motivation. Equinox Publications, ch.2. [abstract] [chapter] [handout]
For stress in particular, see:
de Lacy, Paul (2014). Evaluating evidence for stress systems. In Harry van der Hulst (ed.) Word stress: Theoretical and typological issues, Cambridge University Press, pp. 149-193.
[pdf; Data from EC (docx) (pdf)]
And here’s a shorter reader-friendly article:
de Lacy, Paul (2007). Quality of data in metrical stress theory. Cambridge Extra magazine. Issue 2. [article]