本研究主要探討不同書寫表現之國小學童，其手部精細動作在不同用力程度下，捏力控制的準確性與穩定性相對於不同年級層的比較。實驗分別以國小二、四、六年級各20、20、19名學童為實驗參加者，實驗前由專家教師評定，分為字跡美觀整齊與字跡潦草難認等兩組，實驗要求實驗參加者試做最大捏力之15%、25%以及38%等捏力控制之測試。捏力表現係通過Biometrics E-LINK上肢肌力檢測與分析系統進行資料的收集，並計算絕對誤差（AE）、恆常誤差（CE）、整體變異（TV）、絕對恆常誤差（ACE）、變異誤差（VE）、變異誤差係數（VE-CV）等，相對於不同力量大小的控制進行誤差測量。統計方法採用2（書寫表現）×3（年級）獨立樣本二因子變異數分析，統計的顯著水準訂為α = .05。研究結果顯示，三種力量控制的所有誤差變項在書寫表現與年級二因子的交互作用均無顯著差異（p > .05），年級因子在小、中力量表現也無顯著差異；書寫因子在15%力量的測試中各組力量均超過目標力量，且字跡美觀整齊組的力量控制恆常誤差（0.36 ± 0.37kgw）顯著低於字跡潦草難認組（0.70 ± 0.73kgw）達到顯著差異（F(1,53) = 4.60，p < .05，η2 = .08，power = .56）。在15%、25%捏力控制的書寫表現，VE和VE-CV未達顯著；而38%工作要求下，VE分數與TV分數都是字跡美觀整齊組顯著低於字跡潦草難認組。顯示較小力量控制的準確性與較大力量控制的穩定性對寫字表現仍有影響。

　　This study focuses on the performance of handwriting to compare the grade difference and the control of pinch force for schoolchildren. Fifty-nine health volunteer schoolchildren without learning retardation for each grade group (grade 2, grade 4, and grade 6) served as participants in this study. Participants were selected from two groups, students with good handwriting that could possibly reflect the art of calligraphy and students with poor handwriting that reflects cacography, by experienced elementary school teachers. Every participant was asked to perform relative pinch force in the 15%, 25%, and 38% peak pinch force. The Biometrics E-Link system with pinchmeter was used to record and assess the pinch force between the tip of the thumb and the knuckle of the index finger. The absolute error (AE), constant error (CE), variable error (VE), coefficient of relative variable error (VE-CV), total variability (TV), and absolute constant error (ACE) were used to evaluate the inconsistency and the accuracy of the fine control of pinch force. 2 (the performances of handwritten) ´ 3 (grades) independent two way ANOVA was adopted to analyze the statistical difference with an alpha level of .05. The results showed that all components of force control error were no any statistics interaction between the factors of handwritten performances and grades (p > .05). There were no any grade differences under the small and median control of pinch force. The handwritten factor under 15% peak pinch force showed both groups performed excess of pinch force to achieve the target force, and the constant error for the student with good handwriting (0.36 ± 0.37kgw) was lower than the student with poor handwriting (0.70 ± 0.73kgw; F(1, 53) = 4.60, p < .05, h2 = .08, power = .56). The VE score and the VE-CE score had no any difference between two groups of handwritten under 15% and 25% peak pinch force, but the VE score and the VE-CE score in 38% peak pinch force were showed the student with good handwriting was lower than the student with poor handwriting. It indicated that the accuracy of control in small pinch force and the stability of control in large pinch force affected the handwritten performance.