A review on the problem of incompleteness in sample surveys Yadav Shilpa1,*, Nagar Pankaj2 1Assistant Professor, Department of Statistics, University of Rajasthan, Jaipur 2Professor, Department of Statistics, University of Rajasthan, Jaipur *Email: shilpa.y2108@gmail.com
Online Published on 28 July, 2023. Abstract Conventional survey research procedures rely on Neyman's probability-based sampling paradigm, having a number of fundamental assumptions that are increasingly getting very difficult to achieve in today's survey research environment. One such condition being, an equal and fair chance of every unit of the population to be included in the sample. In most of the studies, complete coverage of the population under study is practically impossible for a variety of reasons. On one hand, where it is difficult to have a perfect sampling frame that is independent of duplication as well as has covered the entire population in the frame, it is also not very economic to construct a fresh sampling frame, every time a study is being conducted. On the other hand, even in the rarest of the cases, if we have succeeded to have a perfect sampling frame, the problem of non-response further creates a void in the study. When cost is another factor that is taken into consideration, it's no surprise that alternative sampling methods are becoming more popular. The authors have taken into consideration, two very common sampling errors, viz., incomplete frame problem and non-response. This study is an attempt to examine and review, a number of past and current practices for drawing conclusions from samples that do not fully adhere to the statistical machinery that is currently available. Top Keywords Incompleteness, Incomplete sampling frame, Imperfect frames, Sampling error. Top |