Basic knowledge for laboratory technicians
Thread Content
Basic Knowledge for Laboratory Technicians Part 1: Basic Knowledge Chapter 1: Allowable Deviations I. Accuracy and Error 1. Accuracy refers to the degree to which the measured result is close to the true value. 2. Error refers to the difference between the measured result and the true value. II. Precision and Bias 1. Precision refers to the degree to which the results obtained in each measurement within the same experiment are close to their average value. 2. Deviation refers to the difference between the measured result and the average value. III. Errors and Deviations Since the “true value” cannot be known with accuracy, it is impossible to calculate errors. In practical work, it is usually the deviation that is calculated (or the error is computed using the average value instead of the true value, with the result still being a deviation). IV. Absolute Deviation and Relative Deviation Absolute Deviation = Measured Value – Average Value Relative Deviation = (Measured Value – Average Value) × 100% If two parallel measurements are taken, with A and B representing the two measured values, then the relative deviation is calculated as follows: Relative Deviation (%) = [(A – Average Value) / (A + B)/2] × 100% = [(A – B) / (A + B)/2] × 100% V. Standard Deviation and Relative Standard Deviation 1. Standard Deviation is a statistical indicator that reflects the dispersion of the measurement values of a set of test samples. If the measured values of the test sample are denoted as Xi, their average is X, and there are n such measured values, then the standard deviation is: 1. Standard Deviation (SD): SX = 2. Relative Standard Deviation (RSD): SX/RSD = ×100%. VI. Maximum Relative Deviation The relative deviation is used to indicate the precision of the measurement results, and a maximum value (also known as the allowable tolerance) is established based on the requirements of the analysis work. VII. Error Limit
Chapter 2 Handling Significant Digits I. Significant Digits 1. The digits that can actually be measured in analytical work are referred to as significant digits. 2. When recording significant figures, it is specified that only the last digit of the number may be imprecise, and the difference from the exact value must be no more than 1. II. Rules for rounding significant figures: Use the \"round up, round down, or keep even if it’s five\" rule to discard excess digits. That is, when the remainder ≤ 4, it is discarded ； If the remainder is ≥6, then include it ； When the units digit is 5, if it is preceded by an even digit, it is discarded; if it is preceded by an odd digit, it is included. When there is any non-zero number after 5, it is included regardless of whether the number before 5 is even or odd. For example: Round the numbers on the left side below to three significant figures: 2.324→2.32, 2.325→2.32, 2.326→2.33, 2.335→2.34, 2.32501→2.33. III. Rules for dealing with significant figures: 1. In addition and subtraction operations, the number of significant figures retained for each number and for their sums or differences is determined by the number of significant figures after the decimal point, using the smallest value as the standard. In addition and subtraction, since it is the transmission of the absolute errors of each value, the absolute error of the result must be equivalent to that of the value with the largest absolute error. For example: 2.0375 + 0.0745 + 39.54 = ? 39.54 has the fewest decimal places; among these three values, it has the largest absolute error, which is ±0.01. Therefore, 39.54 should be used as the reference value. The other two numbers should also retain two decimal places. Thus, the calculation for these three values should be: 2.04 + 0.07 + 39.54 = 41.65. In multiplication and division operations, the number of significant digits in each number, as well as in their products or quotients, is determined by the value with the fewest significant digits. In multiplication and division, since the factor is the transmission of the relative errors of the individual values, the relative error of the result must be equivalent to that of the value with the largest relative error. For example: 13.92×0.0112×1.9723 = ? 0.0112 has three significant figures, the fewest number of digits, and therefore it has the largest relative error; thus, the number of digits in 0.0112 should be taken as the standard. That is: 13.9×0.0112×1.97 = 0.307. The number of decimal places in the analysis results should match the number of decimal places corresponding to the precision of the analysis method. 4. The format of the test results should be in accordance with the provisions of the pharmacopoeia.