For analyzing the difference of forest stand parameters between angle measuring and tallying of square plot,Two datasets were used in this paper,one was composed of 100 temporary square sample plots with the area of 900 m ² measured by tallying and point sampling with angle gauge,another was composed of 1 830 fixed square sample plots with the area of 667 m ² measured by tallying and angle gauge control registration,and then the investigation data of tallying were used as the reference to calculate the error of basal area (G),mean of stand height ($\bar{H}$) and stand volume per hectare (M) of angle measuring accompanied by the pairing t-test. the result indicated that the relative error of G,and M derived from point sampling with angle gauge were -15.88%,5.27% and -12.16% respectively compared with the data derived from tallying,there were more than 50% of total plots,of which relative error of G and M were larger than ±20%,and more than 70% of total plots had a negative error of G and M,the error of G was subject to normal distribution,but that of and M wasn't;there was significant difference (P<0.05) between the mean of G,$\bar{H}$ and M derived from angle measuring and tallying of square plot;the relative error of G measured by angle gauge control registration was -4.90% compared to the that measured by tallying,there were 55.4% of total plots that had the relative error,large than ±20%,and 56.9% of total plots had the negative error;the error of G wasn't subject to normal distribution,the mean of G derived from two method had significant difference (P<0.05) too. All errors of angle measuring were caused by normative and unserious operation,so the quality control of investigation needs to be enhanced to reduce the data error,on the other hand,we need to develop the new technology and new methodology in large-scale forest resources inventory,such as LiDar.