Chapter 4: Statistical Analysis of Random Uncertainties

Random errors cause measurements to scatter around a true value, sometimes yielding overestimates and sometimes underestimates, whereas systematic errors consistently bias results in one direction and cannot be detected through repetition alone. The chapter uses the analogy of target shooting to clarify this distinction: random errors determine scatter among shots while systematic errors determine how far the cluster deviates from the bullseye. The mean of repeated measurements provides the best estimate of the true value, but quantifying measurement reliability requires calculating the standard deviation, which represents the uncertainty of any single measurement and indicates that approximately 68 percent of measurements in a normal distribution fall within one standard deviation of the true value. The standard deviation of the mean, calculated by dividing the standard deviation by the square root of the number of measurements, reveals that repeated measurements reduce uncertainty more efficiently than individual measurements alone, though improvement follows a square root relationship requiring exponentially more trials for linear gains in precision. The chapter emphasizes that combining both error types requires separate treatment or quadrature addition, and recognizes that when results fail to overlap with accepted values, overlooked systematic errors likely warrant investigation. Practical guidance throughout focuses on the sample standard deviation formula using N minus one as the denominator, which provides a more accurate uncertainty estimate for small sample sizes than the population standard deviation formula.