Reliability procedures include Reliability Analysis and Multidimensional Scaling techniques.

Reliability Analysis: This measures the overall consistency of the items that are used to define a scale. As a result, we are given sample size, number of items and reliability coefficients.  

Model- Alpha (Cronbach) is the most popular.  However there are others and the help menu can tell you in what situations each is used.

The “Reliability Analysis” dialog box has one submenu.

Statistics: This will give us additional statistics, such as

	Descriptives for scale if item deleted- Helps us to see if deleting an item will improve our reliability.
	Descriptives for scale- Give you descriptives on the items, as a group.
	Inter-item correlation- Give the correlations between the items.

The data set used for demonstrating the reliability analysis is the Tech Survey data set. See Data Set page for details. The variables of interest are Q31a1 through Q31a12. These variables deal with the level of difficulty of faculty in using technology in the classroom. These questions are rated on a scale of 1 to 5, where 1 is the least difficult to use and 5 is the most difficult to use. Our scale will be the average of these 12 questions.
 

Multidimensional Scaling is a technique for finding pattern in a data set using a distance measure among cases. One should refer to the SPSS Help menu for additional information.