Direct sampling is a somewhat informal term that can refer to when the sample is taken from the actual population and not, for example, some kind of related record like driver’s license’s, voter registration cards, or census forms.
Although the concept sounds simple, direct sampling isn’t always possible. It may be too expensive or time consuming to get a direct sample, or the sampling frame may simply not be available.
As The Opposite of Inverse Sampling
Another definition for a direct sample is where t individuals are randomly drawn from a population. The individuals are tagged and returned to the population. Some time later, a random sample size n is taken, and the number of tagged individuals is counted. In this sense, this is the reverse of Inverse Sampling.
Sampling from Distributions
Direct sampling can also refer to sampling from a particular probability distribution instead of a specified population. For example, in Bayesian theory, sometimes it’s necessary to sample from a posterior distribution, which summarizes what you know after the data has been observed. When you know enough about the posterior distribution, you can directly sample from it. However, in many cases you may not know enough about the distribtuion, or it may be impossible to sample from it (perhaps because of high-dimensionality or complexity).
Non Probabilistic Sampling
Some authors use the term to refer to a particular non-probabilistic sampling method. For example, Bdour & Koury (2009) refer to direct sampling in auditing as where
“Each item sampled is selected based on some judgmental decisions set forth by the auditor who excludes equal opportunity selections and depends more on the intentional selection of items based on criteria which would relate less or more with sample representativeness of the population.”
Bdour, J. & Koury, A. (2009) Statistical sampling in auditing: the case of Jordanian auditor practices. In European Journal of Management. Retrieved December 13, 2017 from: http://www.freepatentsonline.com/article/European-Journal-Management/208535117.html
Dodge, Y. (2006) The Oxford Dictionary of Statistical Terms. Oxford University Press.
Li, X. & Xu, R. (2008) High-Dimensional Data Analysis in Cancer Research. Springer Science & Business Media. Retrieved December 13, 2017 from: https://books.google.com/books?id=ERWwgRM2HWMC&printsec=frontcover#v=onepage&q&f=false
Scheaffer et. al (2011). Elementary Survey Sampling. Cengage Learning.
If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this Statistics with R track.Comments? Need to post a correction? Please post on our Facebook page.