Main Menu

Databases in Disguise

LOCATION: HOME PAGE: DATABASES IN DISGUISE

Statistical Packages

Why it's a Database

Statistical packages such as Minitab, SPSS, and SAS are databases with utilities to perform statistical operations over an entire data set. Like spreadsheets, data sets are generally organized into rows for fields or types of measurements taken and records for each person or object measured. Statistics packages support sorting and some searching, but the primary focus is to perform calculations over the entire set.

Below is a sample data sheet from SPSS showing employee salary, gender, education level and other data sorted by salary.

This page shows the SPSS variable view where specifications for each column or field are indicated. Specifications include field/variable name, field type (numeric or string/text), list of possible values and other specifications. When calculations are done in databases, fields act like variables in many cases.

Why this is important

Statistics packages are excellent for analyzing data, but generally are not designed for long term data storage or data manipulation. The actual data is usually collected somewhere else and imported into a statistics package via a delimited file.

Data collection can take place in Excel, an online survey or be extracted from another database. No matter where the source of the data is, it will be important to set up the file in the statistics package and the delimited file up correctly so that the data can be imported into appropriate fields (columns) and records (rows).

Teaching Ideas

Although you may not wish to devote time to teaching a statistics package, it is important to note that some basic statistics functions can be run in Excel or an online calculator. Some students may find benefits analyzing simple data sets.

Another approach could be to assign students a task of creating and collecting survey data then showing limited tools from Excel for running data analysis.