Table 5.8 contains a paired or diatomic analysis data file, resulting from Segment VI of the DATANAYSIS program. Although 4528 paired combination were characters. To save 4528 characters through the transmission of a special character in place of each encountered pair would require the employment of 138 special characters. Since this quantity is beyond the number of unused and available character sets, one must select a subset to encode. If the 24 most commonly encountered pairs of characters were encoded, this would result in the saving of 1839 characters, providing an additional 1.85 per data reduction. Thus, when combined of 70 per cent can be expected. Finally, Table 5.9 shows the results of segment VII of the program. Is this program segment, the entropy of the data file computed to denote its theoretical statistical compression. Many transmission sequences operate upon 8*N bit boundaries requiring the padding of dummy bits at the end of a word for the transmission of statically encoded data blocks. This will result in the actual statistical compression data reduction approaching but never reaching the theoretically calculated data reduction. The degree of variation will be a function of the transmission block size and the sum of the number of bits per character for each character in encoded block. In is interesting to note that the theoretical 75.77 per cent data reduction is closely approached by the use of the three easily combined data-compression techniques previously discussed. As shown, the execution of this program serves as a valuable guide. Here, it has provided information which indicates that the combination of three fixed-length, easily implemented techniques will approximate the data reduction obtainable from a variable-length technique that requires a lot of processing power and whose use could result in data expansion under certain conditions.