All Times EDT
Keywords: Statistics, Analytics, Binary, Sampling, SAS, SAS Studio, Tasks
SAS is an analytics leader that empowers its customers to turn data into decisions, and SAS Studio Tasks have made it easier than ever for users to run complex statistical analyses using a simple point-and-click interface. The suite of tasks that comes preloaded with SAS Studio is impressive, but customers will always have reporting and analytics needs that cannot be met by these initial tasks. Fortunately, it is possible for SAS users to develop their own custom tasks.
In this poster, I present a custom SAS Studio Task I developed involving sampling plans for binary responses. This task has many industry applications, such as in manufacturing engineering and quality assurance. For example, companies manufacturing medical devices with binary specifications (i.e., Pass/Fail) can use this task to determine the minimum required sample size and maximum number of allowed failures to meet the acceptance criteria for release. The acceptance criteria involve a statistical confidence level and a fixed probability of success (or failure) for each device. SAS Studio Tasks utilize a web-based interface that allows users to provide input values, or options, which are used to update the underlying SAS code. The task can then be run like any other SAS program.
Sample size calculations for experiments with binary responses are trivial for most statisticians to perform. If a statistician is not available, standardized lookup tables are available with some common acceptance criteria plans. However, with my custom SAS Studio Task, a user with no SAS or statistical experience can very quickly determine a sampling strategy for any combination of confidence and probability of success.