構音障礙與音韻障礙為口吃兒童最常被研究之言語障礙，有關說中文口吃兒童之構音能力與音韻能力，則尚無研究加以探討，因此本研究之目的在探討口吃兒童與一般兒童在音韻能力與構音能力的差異情形。研究者以24 名口吃兒童及24名非口吃兒童，共計48 名為研究對象，施測修訂中文口吃嚴重度評估工具（兒童版），篩選出口吃兒童與非口吃兒童，再進行語言樣本分析，之後施測聲韻覺識測驗、音韻覺察測驗、非詞複誦和電腦版構音/音韻異常篩選測驗，二組資料以獨立樣本t 檢定進行統計考驗，探討口吃兒童與一般兒童在音韻能力與構音能力的差異情形。本研究結果發現：口吃兒童與一般兒童僅在音韻能力表現之「聲韻覺識測驗」之聲韻結合分測驗得分、「音韻覺察測驗」總分和「非詞複誦」正確率達顯著差異，皆是一般兒童得分高於口吃兒童，而其他之音韻能力測量二組並無顯著差異；在構音能力表現，二組則無顯著差異，因此本研究的結論口吃兒童似乎在部分音韻能力的表現不如一般兒童，但在構音能力的表現並沒有異於一般兒童。建議在評估兒童口吃時，應將音韻能力列為評估項目之一，尤其是有關聲韻拼合、音韻分辨和非詞複誦的能力。

Purpose: The Covert Repair Hypothesis (CRH) proposed by Postma and Kolk (1993) states that disfluencies reflect the interfering side effects of covert and pre-articulatory repairing of speech programming error during an ongoing speech. Moreover, covert repair occurs during the formulation of the phonetic plan and prior to the translation of articulatory commands into speech movements by the motor system (Postma & Kolk,1993). These researchers suggested that people who stutter have more deficits in their phonetic plans, thus, more occasions for error correction, which result in the breakdowns of their speech; that is, stuttering is a normal reaction of detection and repair to an abnormal phonetic plan and the formulation of speech (Kolk & Postma, 1997). Based on the CRH, children with stuttering have inferior abilities in phonological processing and articulation abilities compared to children without stuttering. Therefore, we compare the phonological and articulation abilities between stuttering and non-stuttering school-age children that speak Mandarin. Methods: All the participants were assessed with the Revised Stuttering Severity Instrument-3 for children (SSI-3, Mandarin Edition), which resulted in 24 pairs of stuttering and non-stuttering children. Furthermore, the measures of Phonological Awareness Inventory (PAI) and Phonological Perception Inventory (PPI) were collected. All children’s Nonword-Repetition (NR) performances including the Accuracy Ratio and Stuttering Syllable Ratio were calculated, and their speech samples were analyzed to find their phonological processes (PP). The correct percentages of consonant, vowel, and tone were obtained with the Articulation/Phonological Disorder Screen Test (A/PDST) to measure articulation abilities. Thereafter, t tests were employed to examine whether differences existed in the collected phonological and articulation measures between the 2 groups. Results/Findings: The results indicated that nonstuttering children had better phonological abilities through the Phoneme Blending assessment of PAI, the Total Score of PPI, and the Accuracy Ratio of NR, than those of the stuttering school-age children. However, no differences appeared in the articulation abilities of the stuttering and non-stuttering groups. School-age children with stuttering may have had inferior performances, compared to children without stuttering, for several phonological abilities including phoneme blending, total discrimination, and phonological working memory. Nevertheless, the 2 groups performed similarly for articulation ability tests. Conclusion/Implications: The etiology of stuttering may originate from the deficits that exist in phonological processes rather than the articulation skills of children with stuttering. We compare the findings of this study to those of previous studies (English) and propose several interpretations. In addition, we discuss the limitations of our study and worthwhile suggestions for future studies.