Sampling is employed in data collection mainly because of the huge cost implications of a census-based data collection, which might be unsustainable and unnecessary for researchers ( Dattalo, 2008 ). Because a sample represents the whole population size, accuracy is of fundamental importance. The sampling method must consider the objectives and reasons for sampling, the relationship between the accuracy of information and precision of data sampling, and the effects of biased estimates. The sampling objectives include examining and producing representative data set estimates of the required parameters with as much accuracy as possible. Sampling also cuts operational costs and the analytical and computing requirements of a survey ( Dunne, 2002 ). The proper sample size is fair in its diversity of parameters such as gender, age, and race.
When interpreted well, precision and accuracy can provide enormous insights into problem areas and possible correct sample situations. Precision considers the variabilities within a sample and provides the level of confidence of the estimates. Simultaneously, accuracy is independent of sample variability and is expressed in percentages to inform the relationship between samples and true population size ( Dunne, 2002 ). Consistent recording of information throughout a research study ensures a good reference for accuracy and precision.
Delegate your assignment to our experts and they will do the rest.
Biased sample estimates may provide values below or above the true population size since they are derived from false information from the sample size ( Palmer and Faloutsos, 2000 ). Faulty data becomes misleading when the users have to use it ignorant of the underlying bias, thus distorting perception and studies. Stratification is the best strategy that can be used in the process of eliminating bias from samples. In stratification, the population of the study is broken down into easier manageable parts.
References
Dattalo, P. (2008). Determining sample size: Balancing power, precision, and practicality . Oxford University Press.
Dunne, M. P. (2002). Sampling considerations. Handbook for conducting research on human sexuality , 85-112.
Palmer, C. R., & Faloutsos, C. (2000, May). Density biased sampling: An improved method for data mining and clustering. In Proceedings of the 2000 ACM SIGMOD international conference on Management of data (pp. 82-92). https://doi.org/10.1145/342009.335384
