The chapters (rather than just the map) give a detailed discussion of how the values are assigned. Note that WALS has some very specific definitions for how certain values are assigned that may not match up with the definitions used by everyone else, although in general they are at least similar to the standard usage, while some specific terminology may differ.
http://wals.info/chapter/16http://wals.info/chapter/17I mostly use WALS for morphosyntactic features and can't comment about Turkish phonology, but one way to check on what's going on is to look at a familiar language.
On Map 16A, I would intuitively expect English to have the same value (lexical) as Turkish does. But it instead it has "Long vowel or coda consonant", which I suppose means that heavy syllables (either with long vowels or with a final consonant) are likely to be stressed. I don't know what the statistics would be on this, but for more information, see the reference given, which I assume discusses this in some detail.
On Map 17A, English has "trochaic" rhythmic stress. This is related to how in English most of the time every other vowel is reduced to a schwa. Turkish doesn't have this pattern or any other one. Trochees are sets of long then short syllables, as is common in English. Iambic stress would be short then long syllables. Turkish has neither of these. In short, (while I may be now overgeneralizing) all syllables in Turkish are given roughly equal weight. Maybe this means that Turkish is a syllable-timed language rather than a stress-timed language, although I've heard phonologists say that those concepts are now outdated and have been (or should be) abandoned.
In general the best way to find out about a specific data point is to follow up with the sources cited below the map (you can scroll through the languages alphabetically or search in the box at the top for the language name). The sources given are usually very good for that language (or the best available), although they are not always cited with page numbers and rarely have any notes explaining why they were interpreted as they were. Still, with a little time you should be able to figure out what's going on in most cases. (If you actually have data or a reference that contradicts what is in WALS you can report an error using the comments system on the website-- see the main page for what the current comments are. But that's not the place to just ask what the values mean.)
