In evaluating an individual's speech reception ability (which could be a social adequacy index) speech audiometry employs standard word lists, phonetically representative of conversational speech sounds. As a preliminary study for the selection of test materials of Korean speech sounds speech confusion test was performed by natural voice of two speakers (one male and one female) against 100 middle school boys and girls whose hearing acuity were normal on selected 100 Korean monosyllable sounds. The results obtained from the above experiments were analized from the E.D.P.S. (Electronic Dater Processing System) to figure out the confusion proportion of each sound, and to observe the confusion tendency, arisen by the arrangement of phonemes (-initial consonants, vowels, and final consonants-) of sounds. The confusion proportion was high in CVCf and CV, and comparatively low in VCf and V. The most influencing phonemes in confusion tendency were ""ㄱ"" ""ㅂ"" ""ㅈ"" ""ㅍ"" ""ㅉ"" ""ㅎ"" in initial consonants, ""ㅏ"" ""ㅜ"" ""ㅣ"" ""ㅓ"" in vowels, and ""ㄴ"" ""ㅇ"" ""ㄷ"" in final consonants.