Objectives: This study aimed to investigate whether there are any differences in pragmatic language skills between preschool children who stutter (hereafter, CWS) and those who do not stutter (hereafter, CWNS) in natural play interactions with their parents. Additionally, the study sought to examine the relationship between pragmatic language skills and children’s stuttering. Methods: 7 CWNS and 7 CWS aged 2-5 years, and their parents participated in the study. In a total of three times at 12-month intervals, their interactions during play activities were collected to analyze abnormal disfluency (AD), normal disfluency (ND) frequency, and pragmatic language skills (initiation utterance of child, initiation of utterance of mother and utterance turn-taking). Between-group comparisons of variables were made by collection period (initial evaluation, 12 months later, 24 months later). Results: First, the AD frequency of the CWS group was statistically significantly higher than that of the CWNS group at the initial evaluation and at 12 months, and there was no significant difference between the two groups at 24 months. Second, the frequency of initiation of utterance in CWNS group was significantly higher than that of the CWS group in the initial evaluation. At 12 months and 24 months, children with stutter had a similar degree of initiation of utterance as that of typical children. Third, there was no significant difference between the CWS and CWNS group in utterance turn-taking. The difference in the frequency of utterances and turn-taking between CWS and CWNS group tended to increase over time. Conclusion: These findings confirmed that stuttering in children is influenced by their pragmatic language skills. This suggests the need for parent-child interaction and parental education programs that enhance children’s pragmatic language skills, which may positively impact fluency in CWS.