The methods described are synthetic estimation, truncated Poisson estimates, multiple-capture surveys in both closed populations (the capture-recapture model and log-linear models) and open populations (the Jolly-Seber model and Markov models), and, more briefly, system dynamics models. The author advises that there is no single best estimation procedures, since the suitability of a method depends on the data that are available and on the nature of the population whose size is to be estimated. He also notes that every technique is based on a model that is an idealization of the process to be described. All involve assumptions that are more stringent than can be shown to hold without qualification. Thus, each involves some error of model specification. A third caution given is that the data on which the estimates are based must be thoroughly understood. Any set of real data contains numerous aspects that are not automatically part of an estimation model. The researcher must look both at the way the data fit the models and at the process by which they were obtained. Finally, the author advises that the application of several of the estimation methods to a given problem yields a better understanding of the phenomenon being studied and a more accurate size estimate than does one method used alone. 6 notes, 9 tables, 6 figures, and 21 references
Downloads
Similar Publications
- The Clinician-Patient Working Alliance: Is It a Significant Predictor of Psychiatric Medication Adherence in a Sample of Recently Released Parolees
- Just Science Podcast: Just the Intersection of Opioids and Illicit Stimulants: Part 2
- Gender Differences in the Effects of Exposure to Intimate Partner Violence on Adolescent Violence and Drug Use