After compiling your data, you may find that you need to do the kind of analysis that only a statistical program can handle. in this special feature we offer evaluations of eight top stats packages-along with advice on how to evaluate others on the market, what those foreign sounding stats terms mean, and seven steps to a successful analysis.
One of the goals of database management is to be able to condense and summarize large volumes of data and make it available for decision making. In pursuit of this goal, database professionals employ a variety of tools, including report writers, query languages (such as SQL and QBE), charting programs, and even spreadsheets. A less commonly used but very powerful tool is the statistical analysis (or simply stat) program.
A stat program is an end-user tool for statistical analysis-a set of measurements and techniques for analyzing and summarizing data.
There are two types of statistical techniques: descriptive and inferential. Descriptive statistics are the most common. They used to describe a dataset and include such measures as the mean (average) and standard deviation. Most database and spreadsheet programs support some descriptive statistics, usually in the form of statistical functions. Stat programs go even further providing more extensive and flexible descriptive methods.
But it's the discipline of inferential statistics where stat programs really shine. The techniques and principles of statistical inference allow us to infer from a given dataset and draw conclusions. (For a more thorough discussion of these terms and others, see the companion article entitled, "A Statistical Primer.") With statistical inference, we can answer questions like:
* Was the increase in …