In key numbered texts you have an added layer not visible to the reader. Someone (or more like a group) other than the translator (definitely someone different in the case of the KJV!) went back and tried to link the English word to a Greek or Hebrew word to some form of the original. That back translation introduces another degree of variability. They attempt to get into the minds of the version translator. Hence they are imperfect back translations of base translations that are also imperfect.
Strong's dictionaries use numbers to identify original language words. Greek is numbered G1 to G5624, Hebrew & Aramaic H1 to H8674. Mounce and Kohlenberger/Mounce use a different numbering system, Goodrick-Kohlenberger, G1 to G6061 and H1 to H10799. Look at these newer dictionaries. They give both the Strong and the G-K numbers.
But the systems, while overlapping, are different. Sometimes the Strong's number is the same as the G-K number, but other times it is not. The newer system has many more numbers.
All the keyed number texts have to pick one of the two systems. That is why (I think; I do not prepare texts for Accordance) each key numbered text is tied to a specific set of dictionaries. KJVS and ESVS are tagged using different numbering systems, Strong's and GK respectively. Accordance's behavior is logical. It is showing you the dictionaries appropriate to the system employed in the back translation.
Edited by Michael J. Bolesta, 21 December 2013 - 05:52 PM.