Statistics Definitions > Ad Hoc Analysis
What is Ad Hoc Analysis?
Ad hoc means “for a particular purpose only.” Ad Hoc Analysis is used in business intelligence to answer a specific question at a specific time; In other words, on an “as needed basis.” The result is a graph, report or other summary that helps you to make a business decision.
Ad hoc analysis can be facilitated with use of a dashboard. A dashboard allows you to get a snapshot view of what’s going on in your area of interest. There’s some debate about the exact definition of what a Dashboard is though. Stephen Few, author of three bestselling dashboard design books, describes a dashboard as:
“…a visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so the information can be monitored at a glance.”
Some dashboards, like the OLAP dashboard, allow you to rapidly access data through a point and click interface.
For big data, Ad Hoc Analysis becomes much more complex. Many businesses use on-site software like IBM’s Hadoop or cloud services like Google Compute Engine, Qubole Data Service or Microsoft Azure.
There isn’t a set definition for what kinds of statistical tests or reports are part of the process. Ad hoc reports are usually generated to fill in gaps left by regular reports, or to further analyze existing reports.
Use in Probability and Statistics
The term Ad Hoc Testing is sometimes used informally in statistics (especially in hypothesis testing) to mean that a test is usable only in very specific conditions, for a very specific purpose. If a researcher uses an ad hoc test, it usually implies that they are using a test in the same way medications are used off-brand. While it may work well for a particular situation, it wasn’t designed for that purpose and shouldn’t be used across the board.
References
Blokdyk, G. (2018). Ad hoc testing Third Edition.
Few, S. Retrieved April 1, 2015 from: http://www.dashboardinsight.com/news/news-articles/interview-with-stephen-few-dashboard-design-challenges.aspx
Vogt, W.P. (2005). Dictionary of Statistics & Methodology: A Nontechnical Guide for the Social Sciences. SAGE.