D.17 STATISTICAL ANALYSIS SYSTEM (SAS)

Approximate Cost: about $3,500 and up, depending on version and add-ins

Source:  SAS (www.sas.com/technologies/analytics/statistics/stat/index.html)

Current Version: v9.3

Operating System Needs: Windows, Unix/Linux

Data Input Structure: Varied, user-defined

Overview

SAS is a suite of statistical software packages that apply to a wide variety of sectors including financial services, medicine, insurance, manufacturing, and oil and gas. Although not designed specifically for environmental applications, SAS is practically unlimited in the types and levels of statistical analyses that it can support. Some backgroundNatural or baseline groundwater quality at a site that can be characterized by upgradient, historical, or sometimes cross-gradient water quality (Unified Guidance). in statistics is required, however, to use all of the features in SAS and to ensure that approaches are valid and inferences are sound.

SAS Base contains many general statistical procedures, as well as graphical capabilities, and is useful for most standard groundwater monitoring applications. More complex matrix-based procedures (such as krigingA weighted moving-average technique to interpolate the data distribution by calculating an area mean at nodes of a grid (Gilbert 1987).) are provided in SAS IML, which is available for separate purchase. Similarly, process control applications (such as Shewhart-Cusum charts) are specifically addressed in SAS QC software, which can also be purchased separately.

Add-Ins Available

Add-in packages are available for a variety of applications and business sectors. SAS/IML and SAS/QC provide added functionality for groundwater monitoring statistics.

Ease of Use and Data Import

SAS version 9.3 (the most recent commercially available version) is available for use on most Windows platforms. SAS is versatile with regard to data file structure, and can read most file types (text files, CSVcomma-separated values, and others), but you generally must define the structure and data format for the program.

Because of its powerSee "statistical power." and versatility, you must understand statistical approaches and methodologies to ensure valid and defensible results from SAS. You can write “free-form” code, calling statistical procedures, and within one program regroup and revise data, perform multi-layered analyses, output new data sets, and provide presentation-grade graphical analyses. SAS can be used to perform reliability analysis.

Visualization

SAS can produce contours, plots, and graphs. For examples of SAS visuals, visit the website http://www.sas.com/technologies/analytics/statistics/stat/index.html.

Primary Uses for Groundwater Data Analysis

SAS performs statistics for a wide variety of applications, but specific methods and tests for groundwater monitoring include statistical intervals, hypothesis testing, regression, correlationAn estimate of the degree to which two sets of variables vary together, with no distinction between dependent and independent variables (USEPA 2013b)., outlier tests, spatial weighting, trend analysis, and a variety of graphical methods. While these procedures are not necessarily available in a drop-down menu or as a specific application, all of them can be coded into an SAS program by a proficient user. User-coded methods include nonparametricStatistical test that does not depend on knowledge of the distribution of the sampled population (Unified Guidance). or non-normal procedures as well as those noted above as available in additional purchased software.

Benefits

• online help and available support system

• many examples provided in the documentation

Limitations and Data Requirements

• cost
• requires at some knowledge of statistical methods

• may be challenging for inexperienced users who may find the initial learning curve steep 

 

Publication Date: December 2013