Legion, I could post just as many contrary articles and please keep in mind
Actually, you couldn't. This was recognized as a serious problem in the 30s, and for several years an annual conference consisting of noteworthy psychologists, physicists, and other scientists met to try to resolve whether measurements in psychology were possible. Although no specific position was reached, the general consensus was "no". The problem is that psychologists didn't have anything to replace measurement theory with (as they applied it), so they just continued to use it. Defenses of measurements in psychology, from Lord and Novick's incredibly influential work to the foundational methods for experimental designs (not only in psychology but sociology, medicine, economics, etc.) founded by Fisher and Neyman-Pearson (which is interesting, as they fundamentally disagreed, yet somehow current NHST is a welding of these contradicting approaches) have either made excuses or otherwise not addressed the criticisms. Thorndike's model was replaced by the largely atheoretical model of Likert mostly for expediency, and any theoretical justification has long since been absent. Item response theory is generally applied the way that statistical methods are, rather than as an empirically, scientifically justified model of measurement (and it failed to actually address criticisms of attitude/personality/etc. measurements in the literature too, as it assumed them). Latent variable theory is worse still, as it is almost entirely approached in terms of the statistical models used to "uncover" latent variables (SEM, path analysis, etc.). For the most part, as with criticisms of NHST, the continued claim that measure of personality, attitudes, etc., in psychology, cognitive neuroscience, social neuroscience, social psychology, etc., are actually measuring anything in any meaningful way persists not due to responses of criticism or even in spite of them, but by ignorance of them.
I studied for my masters in psych with a neuropsychologist.
Fantastic. My field is neuroscience. Most my graduate research was in cognitive neuroscience, and the reason I switched more and more to complex systems, computational neuroscience, neuronal dynamics, physics, nonlinear dynamics, systems sciences, etc., was because of the rampant misuse of scales, statistics, and research designs as well as the unjustified assumptions that most researchers were and are ignorant of because such problems were largely swept under the rug in the earlier half of the 20th century.