The survival probability of freshly nucleated particles governs the influences of new particle formation (NPF) on atmospheric environments and the climate. It characterizes the probability of a particle avoiding being scavenged by the coagulation with pre-existing particles and other scavenging processes before the particle successfully grows up to a certain diameter. Despite its importance, measuring the survival probability has been challenging, which limits the knowledge of particle survival in the atmosphere and results in large uncertainties in predicting the influences of NPF. Here we report the proper methods to retrieve particle survival probability using the measured aerosol size distributions. Using diverse aerosol size distributions from urban Beijing, the Finnish boreal forest, a chamber experiment, and aerosol kinetic simulations, we demonstrate that each method is valid for a different type of aerosol size distribution, whereas misapplying the conventional methods to banana-type NPF events may underestimate the survival probability. Using these methods, we investigate the consistency between the measured survival probability of new particles and the theoretical survival probability against coagulation scavenging predicted using the measured growth rate and coagulation sink. With case-by-case and time- and size-resolved analysis of long-term measurement data from urban Beijing, we find that although both the measured and theoretical survival probabilities are sensitive to uncertainties and variations, they are, on average, consistent with each other for new particles growing from 1.4 (the cluster size) to 100 nm.