This study compared FTIR with XRD method for the analysis of quartz by % recovery, coefficient of variation (CV) and influence of the interference. the results were as the following. 1. In FTIR method, the coefficient of determination ($r^2$) was 0.9998 in a calibration curve of $695;cm^{-1}$, and the limit of detection was $4.9{mu}g/sample$. 2. The highest recovery was $799;cm^{-1}$ (98.2%). 3. The CVpooled of the FTIR method was approximately 10% in three wave numbers. 4. The analysis of qualitative and quantitative for quartz is difficult with mixed cristobalite and iron oxide. 5. In XRD method with rotating sample holder and LynxEye detector, the coefficient of determination was 0.9996 in a calibration curve, and the limit of detection was $5.9{mu}g/sample$. 6. The recovery and CV pooled were 104.3%, and 11 %, respectively. 7. In muffle furnace ashing, the quartz weight decreased to 34% when the maximum weight of the iron oxide was more than eight times. In conclusion, the accuracy (% recovery) and precision (CV) of FTIR and XRD method for analyzing $alpha$-quartz were similar. FTIR method was a disadvantage for sample matrix because it indicates possibility of interference. However, XRD method distinguished specific crystalline forms of silica, and the majority of silicate minerals. In addition, XRD method recommend filter dissolution to pretreatment method.