Recently Orsat Ligorio drew my attention to a fun quick-and-dirty way to measure the development of Latin into the Romance languages. One of the chief problems in charting the evolution of Latin into the various Romance languages is that became the actual spoken Latin that became Spanish, French, Romanian, etc... was rarely written down. Instead, a form of Latin spoken ~200-50 BCE was used.
In other words, people spoke a Vulgar Latin in common usage but wrote in a Classical Latin. This is problematic because our speech dies the second the sound waves dissipate and die, leaving us nothing but the dry bones of an archaic, out-moded tongue. There are rare examples of Vulgar Latin, but it may be said that - broadly speaking - we lack a clear picture of Vulgar Latin into the Romance languages.
But there is a way to tease Classical Latin into divulging its "vulgar" secrets. Angel Lopez-Garcia created the Verbatim Test for Latin. So a speaker of Vulgar Latin may learn Classical Latin vocabulary and even grammar rules, but their inner syntax shapes their preference for things like word order and their taste for particular Classical Latin words over others. A writer would take the Vulgar Latin that they were thinking in and simply translate directly into Classical Latin. We can reverse-engineer that process, however. If we take a text in Classical Latin, we can attempt to translate the text verbatim into a Romance language to see if it makes sense. Then we can see the development of Classical Latin over time as it increasingly fits Spanish, French, Romance, etc... We can also see the breakup of Vulgar Latin into dialect groups (West, East, and Sardinian) as a text that passes the verbatim test for Spanish and French will fail for Romanian, Dalmatian, or Sardinian.
Let's take a look at a work of genuine Classical Latin. Cicero wrote from 80-40 BCE, at the tail end of the era when Classical Latin was still the Latin.
Et tamen te suspicior eisdem rebus quibus me ipsum interdum gravius commoveri... (Cicero, De senectute I.1); "And I suspect, with everything, that sometimes you are seriously worried by the same things that trouble me..." (non-verbative English translation my own)
Attempting to translate this word-for-word into Spanish or French yields only nonsense.
Spanish: Y no obstante te sospecho por las mismas cosas, por las cuales yo mismo a veces fuertemente estar agitado...
French: Pourtant je te doute par les mêmes choses, par lesquelles moi-même parfois gravement être agité...
Now let's look at a work of Classical Latin from much later. St. Jerome will work. He wrote at the tail end of the 300s CE, when Classical Latin had solidified for centuries and Vulgar Latin started its breakup into different dialect clades.
Tunc venerunt duae mulieres meretrices ad regem steteruntque coram eo. (Jerome, Vulgata I Regum 3); "Thus came two women to the king and stood before him." (translation my own)
A verbatim translation into Spanish or French is perfectly intelligible:
Latin [Spanish/French]: Tunc [Entonces/Lors] uenerunt [vinieron/vindret] duae [dos/deux] mulieres [mujeres/femmes] meretrices [meretrices/paillardes] ad [al/au] regem [rey/roy] steteruntque [y se parado/et se mirent] coram [ante/devant] eo [él/luy].
Even though we see many words have been replaced, the overall syntax works. Unlike Cicero, which was nonsensical in Spanish or French. If we wish, we can apply this measure to Latin texts to witness the development of Vulgar Latin and its departure from Classical. Over time, it becomes increasingly difficult to perfectly match the original Classical Latin speakers like Cicero. That's the verbatim test.