<p> Many children, from 0 to 5 years old, spend most of their active hours in child day-care centers. Rooms in these centers are often very noisy due to voices of children and sounds of their various activities, enhanced by the excessive reverberation attributable to the little use of sound-absorbing materials in the room. Some researches have examined the speech intelligibility of elementary school pupils and the results show the score of the lower age group was the more affected by noise and long reverberation. It is expected that pre-school children could be even more affected than low-grade elementary pupils, but no study is found about this. Thus, in this study, experiments were carried out twice in actual classrooms of two child day-care centers with the participation of 3- to 5-year-old children to find the effect of acoustic condition on the word intelligibility of pre-school children.</p><p> The procedure of experiments was like a game so that children can participate in with joy and without losing concentration. After the first experiment (1st Exp.) at a child day-care center, the second experiment (2nd Exp.) was conducted at another center, in which the procedure was revised coping with the problems found in the 1st Exp. Teachers, college students, and elementary school pupils (1st to 3rd grades, 6 to 9 years old) participated in the 1st Exp. and teachers participated in the 2nd Exp. as control groups. Six words that have three morae in Japanese and are familiar to children were chosen for the intelligibility test. Each of the vocalization of the words was mixed with pink noise at one of the three S/N ratios of -5, 0, and 5 dB, and convolved with the room impulse responses of three reverberation times of 0.4, 1.0, and 2.5 s. Children were asked to judge whether the word was food or not and to answer it by raising their hands holding a plate with either yes or no sign on it.</p><p> The correct answers were judged from recorded videos and Correct Answer Score (CAS, %) of each subject was calculated as the ratio of correct answers to the total number of the test words. As a result, it was shown that the mean CAS of the 3- to 5-year-old groups were lower and the standard deviations of CAS of 3- to 5-year-old groups were greater than the teacher group in both of the experiments. On the other hand, these results did not indicate that the score difference between 3- to 5-year-old group and elementary pupil group, and that the cross effects of the age groups and acoustic conditions to the score. Nevertheless, the result that the CAS of 3-year-old group was the lowest as a whole and 34 %, in contrast to 80 % by the teacher group, in the worst condition suggested that pre-school children need a better acoustic environment than adult.</p>