Estimating the Size of Hidden Populations: A Comparison of Four Modern Methods

Abstract

Currently, there exist a number of methods that can be used to estimate the sizes of hidden populations. A hidden population has no defined list of members. Hidden populations often have a negative stigma, such that inclusion in the population is looked down upon or illegal. Examples of hidden populations are people engaged in criminal or deviant behavior such as heroine-addicts, sex workers, or those suffering from an affliction such as HIV. In this thesis, I will explain four modern methods, thereby creating a singular document teaching the reader how to perform each method and the reasoning and logic behind each method, while also discussing their shortcomings. In the first section, I introduce the historical background of a modern method known as the Network Scale-up Method. This method is unique as well as relatively new as its genesis dates back to 1985. An interview with two founders of the method, H. Russell Bernard, and Christopher McCarty, provides background regarding the creation of the method. Unique to my thesis will be the comparison of the four methods. After introducing and explaining the methods, I test each method in an attempt to determine the best technique to use when estimating the size of hidden populations. Little scholarly research has been done in this area, and I hope for my thesis to shed some scientific light on the strengths and weaknesses of each method. Prior to writing this thesis, I knew very little about counting hidden population sizes. Many decisions must be made regarding the dedication of financial, human, and other resources to members of these target populations. Not only is the challenge of 4 making these decisions relevant to health and human services officials, it is also relevant to national spending on crime and a host of other areas. The National Institute of Health alone spent three billion dollars on AIDS research in 2014 (report.nih.gov). Knowing the size of the affected population would not only allow for better allocation of funding, but also would allow for the creation of benefit-programs more targeted to those groups with a higher risk to contract the disease, such as intravenous drug users.