A-weighted SPL was a compromise standard known to give misleading noise results

Evidence

Noise, or unwanted sound, has to be limited. In the 1960ties, there were many methods used to measure noise. Pushed by industry, the International Standardization Organization (ISO) was forced to standardize a method that was not only practical but that would also indicate correct and adequate values. Standardization is a time-consuming process, and it was finally decided to solve the problem in two steps. The first step produced a simple method that could easily be implemented using relatively cheap equipment. The ISO was aware of the fact that this initial method, measurement of A-weighted SPL, might produce inaccurate or even misleading results in noise control. Therefore, the ISO produced a second standard that was not as simple as the former but produced much more appropriate values based on the human sensation of loudness. Two corresponding methods have been described as loudness calculation procedures in ISO 532 and were published only a few years after the dB (A)-proposal. … The method described in ISO 532B is based on what is described in Chap. 8 on loudness and the examples will be elaborated using that method. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1742

Categorization

• XREF: Connects to the broader noise-regulation literature and ISO standard history in One Square Inch of Silence; ISO’s admission that dBA is a practical-but-inaccurate compromise is exactly the kind of regulatory fact that book gestures at.
• QUESTION: Worth chasing: what exactly did ISO acknowledge dBA gets wrong, and how does ISO 532B’s loudness-based method differ in practice?